Alpelisib enhances cisplatin mediated cytotoxicity in non-mutated, PIK3CA-dependent high-grade serous ovarian cancer

Megha Mehrotra; Sourav Chakraborty; Elveera Saldanha; Diksha Joshi; Ajit Dhadve; Asmita Sakpal; Pratik Chandrani; Amita Maheshwari; Santosh Menon; Jaya Ghosh; Sudeep Gupta; Pritha Ray

doi:10.1186/s13048-026-02171-1

. 2026 Jun 23;19:279. doi: 10.1186/s13048-026-02171-1

Alpelisib enhances cisplatin mediated cytotoxicity in non-mutated, PIK3CA-dependent high-grade serous ovarian cancer

Megha Mehrotra ^1,⁶, Sourav Chakraborty ^1,⁶, Elveera Saldanha ^5,⁶, Diksha Joshi ^1,⁶, Ajit Dhadve ¹, Asmita Sakpal ¹, Pratik Chandrani ^5,⁶, Amita Maheshwari ^4,⁶, Santosh Menon ^3,⁶, Jaya Ghosh ^2,⁶, Sudeep Gupta ^2,⁶, Pritha Ray ^1,^6,^✉

PMCID: PMC13543539 PMID: 42332790

Abstract

Background

High-Grade Serous Ovarian Cancer (HGSOC) has a dismal five-year survival rate (< 30%) owing to late detection and recurrence caused by platinum resistance. Targeted therapy exhibits efficacy only within a limited patient population, warranting the need to identify new therapeutic regimes. Herein, we investigated the efficacy of a PIK3CA inhibitor, alpelisib, combined with cisplatin using malignant ascites-derived tumour cells from HGSOC patients, cell lines, patient derived xenografts.

Methods

Malignant ascites-derived tumour cells were collected from treatment-naïve and relapsed HGSOC patients. The efficacy of the cisplatin-alpelisib combination in primary cells and p53 mutant cell lines was assessed through cell viability assays. Targeted next-generation sequencing was performed to identify somatic and germline mutations. Real-time PCR, clonogenic assay, chromatin immunoprecipitation, immunoblotting, immunofluorescence, pre-clinical tumour models, and patient-derived xenograft were used to identify and validate molecular determinants of treatment response.

Results

The cisplatin-alpelisib combination induced significant apoptosis in 33% of patient samples and 80% of the responders had harbored amplification in PIK3CA gene. Intriguingly, primary tumor cells expressing p53 mutations-R196Q or R175H, in a PIK3CA non-mutated background conferred sensitivity to cisplatin-alpelisib. Expression of certain p53 mutants (R196Q, R175H, and R282W) in p53-null SKOV3 cells or EOC cells indigenously harboring p53-R175H (TOV112D) showed enhanced PIK3CA expression following cisplatin treatment. This effect was found to be associated with reduced occupancy of mutant p53, coupled with increased NFκB binding on the PIK3CA promoter, leading to enhanced PIK3CA transcription and susceptibility to the cisplatin-alpelisib combination. The combinatorial treatment effectively arrested TOV112D tumour growth (p53-R175H) but not A2780 (Wtp53) compared to cisplatin alone, and significantly reduced the ascites burden in a PDX mouse model. High somatic BRCA2 mutations were also observed without any association with the cisplatin-alpelisib efficacy in our cohort.

Conclusion

Our findings highlight the potential of cisplatin-alpelisib treatment in a subset of HGSOC patients with PIK3CA overexpression, mediated either through gene amplification or transcriptional modulation, reflecting the wider application of alpelisib in PIK3CA-non-mutated ovarian cancer.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13048-026-02171-1.

Keywords: PIK3CA, Alpelisib, Mutant p53, High Grade Serous Ovarian Cancer, Chemoresistance

Background

High-grade serous ovarian cancer (HGSOC) is the most prevalent and lethal subtype of ovarian cancer, with a 5-year survival rate less than 30%. Patients are often diagnosed at advanced stages and although 75–80% initially respond to first line therapy, majority of them ultimately relapse due to the emergence of drug resistance [1, 2]. HGSOC is the third leading cause of cancer death in Indian women and seventh globally [2–4]. Despite this, platinum and paclitaxel based cytotoxic therapy has remained unchanged for the past two decades. Based on their response to platinum-based drugs and the duration of progression-free survival (PFS), relapsed HGSOC patients are categorised into platinum-refractory (disease progression during treatment), platinum-resistant (recurrence < 6 months from the last chemotherapy), and platinum-sensitive (recurrence > 6–12 months) [5]. Though platinum-sensitive relapse patients respond well to platinum-based drugs, they gradually develop resistance over subsequent relapses. Second-line chemotherapy only shows limited efficacy in platinum resistant cases, which currently have few targeted therapy options [6]. While recent therapeutic advances such as antibody drug conjugate like mirvetuximab soravtansine has shown promising activity in subset of FRα‑positive, platinum‑resistant patients, objective response rate remains limited to 32% [7]. Therefore, it is crucial to identify new therapeutic approaches, especially for the platinum-resistant recurrent HGSOC.

Recently, PARP inhibitors (Olaparib, Rucaparib, and Niraparib) have transformed the treatment of recurrent HGSOC [8]. The high prevalence of homologous recombination deficiency (HRD) (~ 50% of HGSOC cases), including germline and somatic BRCA1/2 mutations (~ 20%) [9], contributes to the success of PARPi therapy. However, the emergence of PARPi resistance remains a significant challenge. Conversely, HR-proficient tumours, often associated with shorter progression free survival, show minimal benefit from PARPi, highlighting the need for developing new targeted therapies. Whole genome analyses of HGSOC identified TP53 as the most prevalently mutated gene (96%), along with mutations present in BRCA1/2, HRR pathway genes, PTEN(7%), NF1(12%), CDK12(4%), and RB1(10%) and other cancer-related genes [10]. The high rate of copy number amplification in genes within the same pathway, like PIK3CA/AKT or NOTCH signalling, drives oncogenic pathway addiction in HGSOC cells. Activating mutations or amplifications in PIK3CA result in hyperactivation of this pathway, promoting cell survival, proliferation, metastasis, and chemoresistance. PIK3CA amplification was also associated with higher AKT phosphorylation [11], and downregulation of AKT was shown to reverse platinum resistance in ovarian cancer cells in vitro [12]. Ohta et al. demonstrated that Wortmannin, a nonclinical pan-PI3K inhibitor, enhanced cisplatin efficacy and reduced ascites formation in an intra-peritoneal xenograft mouse model of CaOV3 (HGSOC cell line) [13]. Recently, various PI3K inhibitors (PI3Ki), including pan-PI3Ki, isoform-selective PI3Ki, and dual PI3K/mTOR inhibitors, have been explored in clinical trials, and some have subsequently been approved. Idelalisib (PI3Kδ) with rituximab or duvelisib (PI3Kδ and γ) are approved for relapsed or refractory Chronic Lymphocytic Leukaemia [14], and alpelisib (PI3Kα) combined with fulvestrant for HR+/HER2- PIK3CA mutant breast cancer [15]. These inhibitors have not yet been approved for any of the subtypes of epithelial ovarian cancers (EOC), including HGSOC. Alpelisib, the only α-specific PI3K inhibitor, is approved for cancers harbouring activating mutations in the kinase and helicase domains. However, utilizing a panel of PIK3CA-altered patient-derived xenografts and cell lines, Fritsch et al. demonstrated that in addition to activating mutations, PIK3CA amplifications are also positive predictors for alpelisib sensitivity [16]. In a recently conducted phase III EPIK-O/ENGOT-OV61 trial, alpelisib-olaparib combination did not show a PFS benefit compared with single-agent chemotherapy of physician’s choice in patients with platinum-resistant/platinum-refractory HGSOC with no BRCA mutation [17]. However, the efficacy of alpelisib in combination with cisplatin, the mainstay chemotherapy for EOC that generates much broader effects than olaparib, has not yet been investigated.

This study explores the combinatorial effects of alpelisib and cisplatin in malignant ascites-derived primary tumor cells from a small cohort of treatment-naïve and relapsed Indian HGSOC patients and subsequently in cell lines and PDX model. Prevalence of somatic BRCA2 mutations, a pathogenic germline mutation in PTCH1, along with certain distinct gene signatures, but no pathogenic mutation in the PIK3CA or PTEN gene, was observed. We showed that PIK3CA copy number amplification is a prime determining factor for enhanced response to cisplatin-alpelisib combination in patient derived primary cells. Several oncogenic p53 mutations (R175H, R282W, R196Q), when expressed in p53-null SKOV3 cells, transcriptionally upregulated PIK3CA through diminished PIK3CA promoter occupancy and increased NFκB recruitment and binding, leading to enhanced sensitivity to combination treatment. In the R175H-bearing TOV112D cell-derived tumor xenograft model, significant reduction in tumor volume by cisplatin-alpelisib combination was observed compared to untreated, and cisplatin treatment. This combinatorial treatment also depleted formation of malignant ascites compared to cisplatin or alpelisib alone in a patient-derived xenograft model harbouring mutant p53-R175H and decreased the viability of tumor spheroids in vitro. Activated PIK3CA/AKT pathway acts as a driver for many cancers, however, so far, use of PIK3CA inhibitors in the clinics is only limited to tumours with mutant PIK3CA and no studies have explored the efficacy of alpelisib and cisplatin in a non-mutated but PIK3CA high setting. This is probably the first report on alpelisib-induced sensitization towards cisplatin for HGSOC, which suggests that upregulated PIK3CA signaling, either due to transcriptional regulation or copy number amplifications, can be used as a predictor of patients’ response to PIK3CA inhibitors.

Materials and methods

Cell lines and culture conditions

SKOV3 (CVCL_0532) and TOV112D (CVCL_3612) cells were cultured in RPMI (Gibco, CA, USA, 11875093); and A2780^LR (CVCL_0134) and IOSE364 (CVCL_5540) in DMEM media (Gibco, CA, USA, 12800017) supplemented with 10% FBS (Himedia, India, RM10409) and 1% Penicillin-streptomycin (Himedia, India, A002), at 37⁰ Celsius and 5% CO2. SKOV3 is a platinum resistant EOC cell line harboring a frameshift mutation in TP53 (p.Ser90Profs*33) resulting in the loss of functional p53 protein. TOV112D is an established EOC cell line that endogenously expresses the p53-R175H hotspot mutation in a wild-type PIK3CA background. The A2780^LR cells were developed by sequentially treating A2780 sensitive cells with incremental doses of cisplatin-paclitaxel [18]. A2780^LR have acquired resistance to cisplatin-paclitaxel and harbour Wtp53. All the cell lines were authenticated by STR profiling and tested for mycoplasma.

Site-directed mutagenesis and construction of mutant p53 clonal library

All TP53 mutations were created by site-directed mutagenesis (SDM) in the Wtp53 construct (pCMV-Neo-Bam-p53, Addgene_16434) using mutation-specific primers. Mutations were confirmed by Sanger sequencing and further stably expressed in p53 null SKOV3 cells using G418 selection. SKOV3-p53 mutant expressing cells were maintained in 300 µg/ml of G418 to ensure stable expression of Wt or mutant p53. SDM Primer sequences are given in Supplementary Table 3.

Lentiviral mediated silencing of NFKB and overexpression

shRNA was designed and cloned into pLL3.7 vector using targeted sequences provided in Table S3. Lentiviral particles were produced as described previously [20]. Briefly, HEK293FT (CVCL_6911) cells were transfected with lentiviral vector plasmid, P-delta packaging plasmid, and VSVG envelope plasmid at a 4:3:2 ratio. Lentivirus particles were harvested at 60 h after transfection. The viral particles were used to transduce TOV112D and SKOV3-R273C cells. Stably transduced cells were sorted for GFP positivity by FACS. For overexpression of NFκB, p65 sequence from pEBB HA RelA plasmid (Addgene_74892) was first cloned into pcDNA3.1 and then transfected into TOV112D^KD and SKOV3-R273C^KD cells. Transfected cells were maintained in 300ug/ml of G418 to ensure stable expression. pEBB HA RelA plasmid was a gift from Dr. Prasanna Venkatraman.

Establishment of primary culture of patient-derived cells

This study was approved by the Institutional Ethics Committee of ACTREC-TMC (IEC#900513). Clinical Trial Number: Not Applicable. Written informed consent was taken from all the patients from whom the samples were collected in this study. Ascites from chemo-naïve, platinum-resistant, and platinum-sensitive HGSOC patients were collected and cultured in MCDB-M199 media supplemented with 10% FBS. After RBC lysis, a dual-step sorting was performed to isolate epithelial tumour cells. First, CD90 + fibroblasts/mesenchymal cells were depleted by Magnetic-Assisted Cell Sorting (MACS, Miltenyi Biotec, USA). The CD90⁻ fraction was stained for EpCAM, and the CD90⁻/EpCAM⁺ fraction was collected by flow cytometry. The CD90⁻/EpCAM⁺ EOC cells were characterised for expression of epithelial lineage-specific marker cytokeratin-18 by immunofluorescence. The cells were grown as a monolayer and all experiments were performed within 2–3 weeks of obtaining the cells.

Genomic DNA Isolation

Genomic DNA was isolated using the QIAamp DNA Mini genomic DNA extraction kit (Qiagen, 51304) as per the manufacturer’s protocol. For the patient-derived tumour cells, 0.5- 1 × 10^6 cells were used for genomic DNA isolation. For the peripheral blood mononuclear cells (Normal DNA), 4-5 ml of whole blood was used for Genomic DNA extraction.

Panel-based NGS for identification of germline and somatic mutations

Panel-based NGS was performed by Medgenome using Illumina HiSeq X, generating 2 × 150 bp sequence reads at ≥ 500x sequencing depth. Fastq files from 14 tumour-normal paired samples were quality-checked with FastQC (SCR_014583) for read count, base quality, and adapter contamination [21]. Low-quality bases and adapters were removed using Trimmomatic v0.39 (SCR_011848) [22]. Reads were aligned to the human reference genome (GRCh38) using the BWA v0.7.17 mem algorithm [23] alignment files were stored in BAM format for further processing and analysis.

Somatic variant calls and filtration

Base quality score recalibration, indel realignment, and duplicate removal were done following the GATK best practices pipeline for short variant and indel calling (v4.2.2.0) [24, 25]. Raw variants were annotated for functionality and pathogenicity using GATK Funcotator [24] and dbNSFP v4.2a [26], which provides pathogenicity predictions from 22 tools. Predictions were coded as 0 (benign) or 1 (pathogenic), and an in-house script calculated their total score. Variants were filtered by the sequencing gene panel, mutation type, variant allele frequency (VAF > 0.01), and dbNSFP pathogenicity sum (≥ 10). Filtered variants were visualized using the oncoPrint function of the ComplexHeatmap package in R v4.2.1 [27].

Germline variant calls and filtration

The BAM files for normal samples were processed to create pileup files using Samtools v1.17 [28] and germline variants were called using the VarScan v2.3.9 mpileup2cns algorithm [29]. Raw variants were annotated with Annovar (2020-06-08) [30]. Pathogenicity scores were calculated from 14 dbNSFP prediction tools. Variants were filtered based on mutation type, panel genes, and a pathogenicity sum (dbNSFP sum ≥ 7). Filtered variants were visualised using the oncoPrint function of the ComplexHeatmap package in R v4.2.1 [27].

Copy number assessment

PIK3CA copy number was assessed using genomic DNA isolated from peripheral blood mononuclear cells (PBMC), CD90-/EpCAM+ cancer cells and PDX cells via the comparative CT method using Real-Time PCR. Estimation of copy number involved amplifying PIK3CA and a reference gene with a known copy number. Three reference genes were used for normalisation: GAPDH, ACTB, and COL8A1 (present on the same chromosomal arm as PIK3CA). GAPDH from the IOSE364 cell line (human immortalised ovarian surface epithelial cells) was also used as an additional reference control. For assessment of the copy number of P16 and PDX-P16, GAPDH, ACTB, and COL8A1 from the IOSE364 cell line were used as a reference control. Relative copy number was calculated using the formula: Copy number = 2^-[(test-ref)_TC - (test-ref)_normal], where Test- PIK3CA, TC- Tumour cell; Ref-Reference control; Norm- Paired normal. Primer sequences are included in the Supplementary Table 3.

Quantitative real-time PCR

RNA from untreated and cisplatin-treated cells was extracted using the RNeasy Mini Kit (Qiagen, 74104). cDNA was prepared using SuperScript IV (18090010, Invitrogen, USA). Quantitative real-time PCR was performed with PowerUp SYBR Green Master Mix (A25741, Invitrogen, USA) using gene-specific primers and GAPDH as an internal control. The relative gene expression was measured as delta Ct values. The list of primer sequences is included in the supplementary Table 3.

Gene expression profiling using PCR array

Gene expression profiling was performed using the RT² Profiler PCR Array (Qiagen_330231) according to the manufacturer’s instructions with minor modifications. Total RNA was reverse transcribed using the PrimeScript RT Reagent Kit (Takara Bio_RR037A) to generate cDNA. The cDNA was then combined with RT² SYBR Green ROX qPCR Mastermix (Qiagen_330520) and 10 ul was dispensed into the RT² Profiler PCR Array plate. Real-time PCR was performed on an QuantStudio5 instrument. Raw Ct values were exported and uploaded to the Qiagen GeneGlobe Data Analysis Center (https://geneglobe.qiagen.com), where fold-change calculations and volcano plot generation were performed using the ΔΔCt method by normalizing with the housekeeping genes. Genes with a fold-change greater than 2 and a p-value ≤ 0.05 were considered significantly differentially expressed. Pathway enrichment analysis of the significantly upregulated and downregulated gene sets was subsequently performed using the WEB-based Gene Set Analysis Toolkit (WebGestalt; https://www.webgestalt.org) with the Reactome pathway database.

Immunoblotting

Cell lysates were prepared after trypsinization using Laemmli lysis buffer containing protease and phosphatase inhibitors. 40-60ug of proteins were resolved on 8% denaturing polyacrylamide gels, then transferred onto PVDF membrane (PALL, NY, USA, 741260) followed by blocking with either 5% BSA or milk in TBST (0.1% Tween) as per the antibody protocol. The blots were then probed with the respective primary antibodies and secondary antibodies and developed as described previously [20]. The blots were quantified by densitometric analysis using Biorad ImageLab Software. Densitometric values for each lane were normalized to their respective loading controls and are expressed as fold-change relative to untreated samples and presented under each blot. pAKT levels were further normalized to their respective total AKT. The list of antibodies is included in Supplementary Table 4.

Nuclear and cytoplasmic fractionation

Nuclear and cytoplasmic fractionation was performed as described previously, with slight modifications [31]. 2 × 10⁶ cells were lysed with 200ul (five times the cells pellet volume) of Cytoplasmic Extract (CE) buffer [(10 mM KCl, 0.1 mM EDTA, 10 mM HEPES and 0.3% NP-40 with protease inhibitors at 4⁰ C for 30 min with intermittent gentle vortexing. The lysate was centrifuged for 20 min at 6000 rpm at 4⁰ C and supernatant was collected as the cytosolic extract. The nuclear pellet was washed thrice with CE buffer by centrifugation. The pellet was resuspended in one volume of ice-cold nuclear extraction buffer (20 mM HEPES (pH 7.9), 420 mM NaCl, 20% glycerol, 1 mM EDTA and proteinase inhibitors) and incubated on ice for 30 min, with intermittent vortexing. Both nuclear and cytoplasmic lysates were further sonicated and lysates were centrifuged for 20 min at 14,000 rpm at 4⁰ C , and the supernatants were collected as nuclear and cytoplasmic extracts, respectively. Immnunoblot analysis was performed as described before. Lamin A/C and α-tubulin were used as the nuclear loading control and cytoplasmic loading controls respectively. The list of antibodies is included in Supplementary Table 4.

Immunofluorescence

Cells seeded on coverslips were fixed in 4% paraformaldehyde and permeabilized with 0.2% Triton-X100. Cells were blocked with 3% BSA and co-stained with both p53 and NFκB primary antibodies at 4⁰ C, overnight. After three washes the cells were probed with their respective secondary antibodies at room temperature for 2 h. Coverslips were mounted in Vectashield Antifade Mounting Medium (Vector Labs, H-1000–10, Vector Laboratories, Burlingame, CA, USA), and imaged in the Nikon AX Confocal Microscope. Colocalization analysis was performed using the ComDet plugin in ImageJ-Fiji. A minimum of 50 cells per condition was counted. The list of antibodies is included in Supplementary Table 4.

PIK3CA promoter analysis

The 934 bp PIK3CA Promoter sequence was adapted from Gaikwad et al., 2013 [19]. Transcription Factor binding sites were identified using the Alggen Promo Database at 5% dissimilarity rate [32]. The promoter schematic with overlapping TF binding sites was created using SnapGene Viewer software v7.2.

Chromatin immunoprecipitation

ChIP assay [33]was performed using Micrococcal Nuclease (Thermo Fisher- 88216) digestion, as per the manufacturer’s protocol. Briefly, 1 × 10⁶ untreated and cisplatin-treated cells were crosslinked with 0.75% formaldehyde for 10 min at room temperature, quenched with 125mM ice-cold glycine for 5 min and lysed to obtain the nuclear fraction. MNase digestion (0.5 µL, 37 °C, 7 min) and mild sonication produced 150–300 bp fragments. Chromatin (25 µg) was immunoprecipitated with ChIP-grade p53 antibody (2 µg), enriched using Dynabeads™-Protein G (Invitrogen™, MA, USA, 10003D), and analyzed by Real-time PCR using site-specific primers (Supplementary Table 3). Non-immunoprecipitated chromatin served as the input control.

Cell viability assay

MTT (Sigma-Aldrich, M2128) was used to assess the viability of cells treated with cisplatin (Sigma-Aldrich, P4394) and/or alpelisib (Medchemexpress, HY-15244). For all combination treatment experiments, cells were treated with their respective IC₂₀-IC₃₀ does of cisplatin and alpelisib, alone and in combination. For SKOV3, TOV112D, and A2780^LR cells, 1 × 10³ cells/well were seeded, treated with various concentrations of cisplatin and/or alpelisib for 72 h, followed by incubation with 10% MTT solution for 2 h and addition of 100 µL DMSO. Absorbance was measured at 560 nm with background correction at 670 nm. Cell viability was calculated as [Absorbance (Test) / Absorbance (Control)] × 100. For patient-derived and PDX cells, 2 × 10³ cells/well were seeded, treated with cisplatin and/or alpelisib for 48 h and 72 h, respectively, and then incubated with 10% Alamar for 8 h. Fluorescence was measured at excitation 530 nm and emission 590 nm. Cell viability was calculated as [Fluorescence (Test) / Fluorescence (Control)] × 100].

Long-term survival assay

Clonogenic assay was performed as described previously [31]. Briefly, 500 cells per well were seeded in a 6-well plate. Cells were allowed to adhere and then treated with cisplatin (0.5–1.5 μm) for 24 h. The plates were further incubated for another 7 days. Colonies were stained with 0.05% crystal violet and counted using a stereomicroscope.

Detection of apoptosis by annexin PI

For detection of apoptosis, TOV112D cells were stained with Annexin V-conjugated FITC and propidium iodide (PI) using the Dead Cell Apoptosis Kit (Invitrogen™, MA, USA, V13242), as per the manufacturer’s protocol. Briefly, control and treated cells were harvested and then resuspended in 100 µl of Annexin V binding buffer, followed by adding 5 µl of Annexin V-FITC and 5 µl of PI, and incubation for 15 min in the dark. After incubation, 400 µl of Annexin V binding buffer was added to each tube and analysed (50,000 events) using an Attune NxT Flow cytometer (ThermoFisher, MA, USA). Flow cytometry data were analysed using FlowJo version 10 software.

Boyden chamber assay

4 × 10⁴ R196Q and Wtp53 cells suspended in incomplete (without FBS) medium were stained with 0.5µM Cell Trace CFSE (Thermofisher Scientific, C34554) for 30 min and seeded in the matrigel-coated Transwell inserts, which were then placed in 400 µl of complete medium and incubated at 37^◦C for 16 h. After incubation, the Matrigel (Corning, 354234) was removed from the insert, and the insert was mounted on the slide. The insert membrane was further imaged for quantification using Olympus BX63F Microscope.

Wound-healing assay

Wtp53 and R196Q cells were seeded in 6-well plates to obtain monolayers. Cells were treated with 4 µg/ml of MitomycinC for 2 h to prevent cell division. Post-MitomycinC treatment, the cells were thoroughly washed with PBS, and a wound was created using a sterile pipette tip. The closure of the wound was then observed and imaged in real time over 24 h using the Sartorius Incucyte Live-Cell Analysis System (SCR_023147).

Tumour xenograft studies and bioluminescence imaging

All animals used in this study were provided by the Animal House Facility of ACTREC-TMC. All the animal studies were performed per the approval of the Institutional Animal Ethics Committee of ACTREC (IAEC#26/2020). 5 × 10⁶ luciferase-tagged SKOV3 cells expressing R196Q or R273H were injected subcutaneously into the flanks of 4–6-week-old female NOD-SCID mice(n = 3). Bioluminescence imaging was performed using 30 mg/kg D-luciferin (intraperitoneal injection) on the Xenogen-IVIS, and image quantification was conducted using Living Image software v4.4.

To assess the in vivo efficacy of cisplatin and alpelisib, 5 × 10⁶ TOV112D cells or A2780^LR cells were implanted subcutaneously in the flank of NOD-SCID mice. Mice were anesthetized using Isoflurane as per the standard procedures. Once palpable tumours appeared, the mice were randomly segregated into four groups: vehicle-control(n = 7), cisplatin(n = 6), alpelisib (n = 5), and combination (cisplatin + alpelisib) (n = 6). For A2780^LR, the groups were as follows: vehicle-control (n = 6), cisplatin(n = 6), alpelisib (n = 6), and combination (cisplatin + alpelisib) (n = 6). alpelisib was administered at 25 mg/kg (10% DMSO in corn oil) daily by oral gavage for 7 days (days 15–21 post-implantation for TOV112D and Days 16–22 for A2780^LR). Cisplatin (4 mg/kg in saline) was administered twice: days 14 and 19, and days 15 and 20 for TOV112D and A2780^LR, respectively. The control groups received appropriate vehicle (10% DMSO in corn oil) solutions. Tumour growth was monitored by measuring the tumour volume. To avoid probable PIK3CA inhibitor-mediated toxicity, such as hyperglycemia, diarrhea, etc., the treatment regimen was concluded on days 21 and 23 for TOV112D and A2780^LR, respectively. Tumor growth for some mice was monitored for an additional 10 days. Mice were euthanized on days 31 and 33 for TOV112D and A2780^LR, respectively, and tumors were harvested for histopathological analysis. Tumor volume was calculated using the formula (V = (length × width²) / 2). All mice were euthanized by carbon dioxide inhalation. 100% CO2 was gradually released into their cages for 2–3 min, till death was confirmed by a veterinarian. The tumors were paraffin-embedded, sectioned, and stained with hematoxylin and eosin. A pathologist conducted blinded histopathological evaluations of two sections per tumor to assess the percentage of necrosis. Images were captured with the 20X objective using the Zeiss Axio Imager.Z1 microscope.

Establishment of patient-derived xenograft

Ascitic fluid collected from a chemotherapy-naïve HGSOC patient enriched in spheroids was resuspended in complete MCDB: M199 medium, and 100 µL of the suspension was intraperitoneally injected into 6 to 8-week-old NOD SCID mice. Mice were monitored for ascites development, which typically occurred within 60 days. Upon development of ascites, mice were sacrificed and ascitic fluid was harvested, and 100ul of ascites was reinjected into new mice for subsequent propagation of PDX cells. All mice were euthanized by carbon dioxide inhalation. 100% CO2 was gradually released into their cages for 2–3 min, till death was confirmed by a veterinarian.

The development of ascites in the mice was tracked by measuring the changes in body weight. For the assessment of in vivo efficacy of the combination treatment, mice at day 60 were randomly segregated into 4 groups (n = 3), and treated with cisplatin, alpelisib, or a combination of both. Cisplatin (4 mg/kg body weight) was given on days 60 and 65, while alpelisib (25 mg/kg) was given daily for 8 days (61–68). Tumor spheroids harvested from malignant ascites were assessed ex vivo for drug sensitivity. Isolated spheroids were plated in 12-well plates and treated with cisplatin, alpelisib alone, and in combination at IC₂₀ concentrations for three days. The remaining spheroids were imaged using an Olympus DP23 Camera with an Olympus IX51 microscope using the 4X objective, and their size and viability were quantified using the ImageJ Analyse Particles tool, following threshold adjustments to define spheroid boundaries. Data was analysed to assess treatment response across 72 h. The spheroids were also allowed to adhere for measuring drug efficacies and PI3K/AKT pathway deregulation.

TCGA data analysis

We accessed and analyzed ovarian serous cystadenocarcinoma data from the TCGA Firehose Legacy dataset (n = 310) using the cBioPortal platform (https://www.cbioportal.org). PIK3CA alterations, including mRNA expression levels (U133 microarray), copy number amplifications, and somatic mutations, were evaluated across all available samples. To assess the impact of TP53 mutations on PIK3CA expression, patients were segregated based on the type of missense TP53 mutations they harbour. The median PIK3CA mRNA levels were calculated for these subgroups and compared to the median PIK3CA mRNA levels of patients harbouring wild-type TP53. TP53 mutations present in fewer than four patients were excluded from the analysis.

Statistical analysis

For majority of the experiments (MTT, western, RT-PCR, etc.) presented here as the mean ± SEM consisted of at least three independent biological replicates. For ChIP, analysis, n = 2 biological replicates are presented. For patient-derived experiments, technical replicates are reported for individual patient samples. Unpaired two-tailed t-test was used for pairwise comparisons (cell viability, RT-qPCR, ChIP-qPCR). One-way ANOVA was used for immunofluorescence colocalization analysis across multiple groups. Two-way ANOVA with Sidak’s multiple comparisons correction was applied for tumor volume measurements at the end of the treatment period and at the end of the holiday period. For PDX body weight comparisons across treatment groups in xenograft experiments, one-way ANOVA was used. Unpaired t-test was used for the tumor necrosis analysis. p-value ≤ 0.05 was considered as significant.

Results

PIK3CA amplified primary tumor cells exhibit sensitivity to cisplatin and alpelisib combination

Malignant ascites, a characteristic symptom of HGSOC patients during disease progression, often reappears in recurrent patients. CD90⁻/EpCAM⁺ epithelial tumour cells were collected from malignant ascites of 15 advanced-stage HGSOC patients, comprising chemonaive (n = 8), platinum-sensitive relapse (n = 4), and platinum-resistant relapse (n = 3) patients (Table S1) and tested for their response towards cisplatin and alpelisib, alone and in combination, at respective IC₂₀-IC₃₀ doses. Patients exhibiting > 15% decrease in cell viability to the cisplatin-alpelisib combination, compared to monotherapies, were classified as Responders whereas those showing no significant decrease were categorized as Non-Responders (Fig. 1A). The patient derived cells showed distinct responses to cisplatin but not much to alpelisib. (Fig. 1B, C). Alpelisib marginally decreased cell viability (Fig.S1 A, B), with IC₅₀ values not reached in all patients; thus, IC₂₀ doses were extrapolated using GraphPad Prism. No association was found between cisplatin response and alpelisib sensitivity.

Fig. 1 — *PIK3CA* amplified primary tumor cells exhibit sensitivity towards cisplatin-alpelisib combinatorial treatment: (A) Schematic showing stratification of patients into responders and non-responders based on their response to cisplatin-alpelisib combination. (B-C) Cell Viability assay using malignant ascites derived tumor cells showed differential cisplatin response by (B) Responder-Patients (Patients P2, P5, P6, P10, P14) and (C) Non-Responder- Patients (P1, P3, P4, P7, P8, P9, P11, P12, P13, P15). Here response to cisplatin-alpelisib combination was used to classify the patients into Responders and Non-Responders. (D, F, H, J, L, N) Graphical representation of cellular viability for cisplatin, alpelisib, and combination treatment at IC₂₀-IC₃₀ doses for Responders (D) Patient 2, (F) Patient 5, (H) Patient 6, (J) Patient 10, (L) Patient 14, and Non-Responder(N) Patient 8. Data represented as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, n.s.- non-significant, as calculated by unpaired TTest. (E, G, I, K, M, O) Immunoblots of patient-derived cells treated with cisplatin, alpelisib, or combination treatment showing changes in levels of pAKT, p110α, p53, and Cleaved PARP for (E) Patient 2, (G) Patient 5, (I) Patient 6, (K) Patient 10, (M) Patient 14, (O) Patient 8. (P) Graph showing relative *PIK3CA* copy number in Patient-derived cells normalized with *GAPDH* of respective PBMCs (Peripheral Blood Mononuclear Cells). Relative Copy Number was calculated using the Formula Copy number = 2^-[(test-ref)TC - (test-ref)normal], where Test- *PIK3CA*, TC- Tumour cell; Ref- Reference control; Norm- Paired normal. Each graph represents three technical replicates. (Q) Nested pie chart depicting the association of *PIK3CA* transcript levels with *PIK3CA* genomic alterations in the TCGA Firehose legacy Dataset (n = 310). The Outer ring depicts changes at the mRNA level. The inner ring depicts changes at the genomic level. The pie chart was created using Microsoft Excel. P denotes patient

While cisplatin treatment induced differential AKT phosphorylation, a reduction in phospho-AKT by alpelisib, an effect of PIK3CA inhibition, was observed for all cases (Fig. 1E, G,I, K,M, O & Fig. S2). Intriguingly, combinatorial treatment of cisplatin-alpelisib exhibited decreased viability (63%,21%, 40%, 27.5%, and 30%) in 5 out of 15 patient-derived cells (P2, P5, P6, P10, P14), respectively (Fig. 1D, F,H, J,L). Due to the paucity of tumour cells or early senescence in some cases, the sole effect of alpelisib could not be tested for all. Increased PARP cleavage was seen with the combination or cisplatin treatment, but not with alpelisib alone (Fig. 1E, G,I, K,M, ). Non-responders did not show decreased cell viability (Fig. 1N, Fig.S2) or PARP cleavage with combination treatment (Fig. 1O and Fig.S2 B, D,F, H,J, L,N, P,R).

Since no mutations or deletions in PIK3CA and PTEN genes were observed, somatic copy number amplification (CNA) was assessed for PIK3CA by comparing it with corresponding peripheral blood mononuclear cells. Chromosomal instability, gene duplication, and deletions are well-recognized aberrations in HGSOC cells. Thus, two housekeeping genes (GAPDH and ACTB) and a genomic-locus-specific gene, COL8A1 (3q12.1), were used for normalization. Additionally, the PIK3CA copy number of patient cells was normalised to an immortalised ovarian surface epithelium (IOSE) cell line (Table 1). Four cases, except P2, showed amplification (relative copy number > 2) compared to the reference genes (Fig. 1P). Intriguingly, P2, without any PIK3CA CNA, exhibited maximal cell death (45.74%) with cisplatin-alpelisib (Fig. 1D). Overall, ~ 26% (4 out of 15) of patient-derived tumor cells showed PIK3CA amplifications and response to combination treatment, with none in the non-responder group (10 out of 15) (Fig. 1P). TCGA analysis of HGSOC patients revealed no PIK3CA alteration in 64.5%, mutations in only 0.6%, gene amplification in 28.7% of cases, and exclusive transcript level upregulation without any genomic alterations in 6.8% of cases (Fig. 1Q).

Table 1.

Relative PIK3CA copy number determined for primary tumor cells of the responders

		PIK3CA Copy Number (Relative to Reference Gene) Copy number = 2^-[(test-ref)_TC - (test-ref)_normal]
	Reference Gene	GAPDH(PBMC)^a	ACTB(PBMC)^b	COL8A1(PBMC)^c	GAPDH(IOSE)^d
Sample Number	P5	2.56	3.26	2.08	3.54
	P6	4.31	2.35	3.35	2.53
	P10	16.42	6.46	0.10	20.26
	P14	8.34	4.22	4.98	3.69

Open in a new tab

Relative copy number of PIK3CA gene normalized to the copy number of (a) GAPDH from PBMCs. (b) ACTB from PBMCs. (c) COL8A1 from PBMCs and (d) GAPDH from Immortalized Ovarian Surface epithelial cells (IOSE). PBMC denotes genomic DNA isolated from Peripheral Blood Mononuclear cells of each individual patient

Mutational spectrum of primary tumour cells

Targeted genomic sequencing using a 147-gene panel was performed on 14 out of 15 samples (one was excluded due to insufficient DNA yield). For P16, the source for the PDX model, we were not able to collect blood and thus could not assess the germline mutational spectra. TP53 mutations, including missense, nonsense, splice variants, and intronic mutations, were found in 86% of cases (Fig. 2A). P1 harboured a 5’UTR mutation in WRAP53, which encodes an antisense RNA that stabilises p53 mRNA (Fig. 2B), resulting in a p53-null phenotype despite an unaltered TP53 gene. P11 and P14 also exhibited a null-type phenotype, confirmed by immunoblotting (Fig. S1C) and immunohistochemistry (IHC) obtained from the electronic medical records of the patients, NGS revealed splice site insertions and deletions in P11, while P14 harbored a frameshift deletion in exon 8 (Fig. 2B). Pathogenic missense mutations in p53 DNA binding domain (exons 5, 6, and 8) were observed in 35.7% of patients (P2, P3, P4, P7, P15). Two novel TP53 mutations, R196Q and Q165L, previously not reported for ovarian cancer, were identified in P2 and P7(Fig. 2B). 80% of mutations were homozygous except for P3 and P7 (Table S2, Fig. 2B), with P3 having two concurrent mutations (R273C, R283L). Sanger sequencing confirmed all somatic DNA binding domain mutations, except R196Q, possibly due to its low allelic frequency (3.4%). Somatic intronic mutations (insertions, deletions, and missense) of unknown significance were observed in 71.4% of the patients (Fig. 2A), which may potentially affect protein splicing, resulting in the translation of a defective p53 protein. A heterozygous germline mutation (P72R), associated with breast cancer [55], was present in 10 out of 15 patients (Table S2, Fig. 2B).

Fig. 2 — Mutational spectrum of Primary Tumour cells: (A) Oncoprint plot showing somatic pathogenic mutations for frequently mutated genes in HGSOC patients of the cohort. Mutations are annotated as Splice(Splice site mutations), Trunc(Frameshift mutations and truncations), Missense-(Missense mutations), Inframe (In-frame insertions and Deletions), and Intron(Intronic Mutations(Shown only for *TP53*). (B) Table showing pathogenic mutations in the DNA binding domain of TP53(*Mutation in *WRAP53*, a *TP53*-associated gene, renders a p53 null phenotype) and (C) Oncoprint plot showing germline Pathogenic mutations in the patient-derived cells. (D) Schematic showing the clinical status of the responders and non-responders since sample collection with a median follow-up of 31 months. (E) Oncoprint plot showing pathogenic *BRCA1* and *BRCA2* mutations in patient-derived cells. For Oncoprint plots (A, C, E), percentages on the left Y-axis indicate the mutation frequency of each gene in the cohort. The right Y-axis denotes the gene names and corresponding mutation count. The horizontal bars on top indicate the mutation count per sample

Low-frequency somatic mutations in BRCA1/2 and HGSOC-associated genes including CHEK2, RB1, ATM, CDK12, BRAF, and NF1 were observed in the cohort (Fig. 2A) along with several alterations in some uncommon genes (Fig.S3). Additionally, an in-frame deletion in RET (at leucine16) with low allelic frequency (1–8%) was observed in 64% of cases, along with frameshift deletions in PTCH1 (14%), mutations in AKT1 (21%), and ATRX (29%). While truncating mutations were observed in ALK and AR genes, a novel, putative pathogenic somatic mutation-R586L in FGFR1 was identified in two (P1 & P9) of the three patients with altered FGFR1 (Fig.S3). P13, harboring a high mutational burden, had was found to possess POLD1 and POLE mutations (Fig.S3), known to be linked with a hypermutant phenotype in colorectal and endometrial cancers [34–36]. Pathogenic germline mutations in homologous recombination related genes were observed in 7% to 29% of cases (Fig. 2C). Clinical outcome and survival were followed for ~ 31 months(median) (Fig. 2D). Germline BRCA1 (21%) and BRCA2 (14%) alterations, primarily pathogenic frameshift deletions, were observed in our cohort (Fig. 2E). BRCA1 mutations included frameshift deletions in exons 14 (P1) and 11 (P5), and two missense variants of unknown significance (VUS): T176M (exon 7) and E848D (exon 11) in P5. For BRCA2, a frameshift deletion in exon 10 (P3) and a missense VUS (A1996T) in P6 were identified. All three VUS were classified as potentially pathogenic by seven out of fourteen dbNSFP prediction tools. Contrary to Western populations, somatic low-frequency frameshift insertions and deletions (1.1–5.5%) were found in BRCA2 (Fig. 2E) in 43% of the patients. P13 showed significant clonal variation with seven distinct BRCA alterations at low allelic frequencies (1.7–3.3%), including deletions in BRCA2 exon 10 and BRCA1. The somatic BRCA2 truncations were also higher in the responders (75%) than in non-responders (20%) (Fig. 2E).

Transcriptional and functional characterization of the novel mutant R196Q

The novel R196Q TP53 mutation, identified for the first time in HGSOC, was functionally characterized in SKOV3 cells. R196Q, is a rare, partially functional p53 mutation found in ~ 0.04% of all cancers, including lymphoma, bladder, and breast cancers [37–40]. To assess the functional contribution of this mutant, we compared the classical p53 target gene expression between Wtp53 and R196Q using a p53 specific PCR array kit. Volcano plot analysis revealed that classical Wtp53 transcriptional targets, including key mediators of cell cycle arrest (CDKN2A, BTG2, CCNG1), apoptosis (BAX, BBC3/PUMA, FAS, CASP9), and DNA damage response (MDM2, CHEK2, SESN2), were downregulated in R196Q-expressing cells (Fig. 3A, B). R196Q also conferred a distinct transcriptional signature characterized by upregulation of key oncogenic genes, most prominently EGFR (~ 19-fold), JUN (~ 24-fold), IL6 (~ 7.5-fold), E2F1 (~ 7-fold), and also upregulation of KRAS, IGF1R, BCL2A1, and BRCA1 (Fig. 3A, B). Pathway analysis of the significant up or downregulated genes revealed that R196Q upregulates genes enriched in EGFR transactivation and ERBB2 signalling, while simultaneously suppressing canonical TP53-driven apoptosis pathways, including the intrinsic apoptosis pathway and transcription of cell death genes, and cell cycle arrest program (Fig. 3C, D).

Next, we performed qPCR analysis for selective genes observed to be significantly altered in PCR array along with p21, p27, and NFκB, following cisplatin treatment in SKOV3-R196Q and SKOV3-Wtp53 expressing cells (Fig. 3E). We observed that Wtp53 induces canonical tumour suppressor targets, particularly p21, PUMA and BAX. In contrast, in R196Q, p21 induction is significantly diminished and NFkB expression was observed to increase. EGFR fold change with cisplatin treatment also increased 20-fold, representing the highest expression across the entire gene panel (Fig. 3E).

Functionally, SKOV3-R196Q displayed rapid wound closure (Fig. 3F, G) but similar invasive properties compared to Wtp53 (Fig. 3H, I). Cell cycle analysis revealed the absence of S-phase arrest in R196Q cells in response to cisplatin, a response that was observed in Wtp53-expressing cells (Fig. 3J). Consistent with this, γH2AX levels were comparatively lower in R196Q cells upon cisplatin treatment (Fig. 3K), suggesting an attenuated DNA damage response. These properties reflect the distinct transcriptional signature of this mutant. Intriguingly, R196Q-SKOV3 cells did not generate tumors in NOD-SCID mice, while R273H-SKOV3 cells formed tumors and sustained bioluminescence in the same mice for 29 days (Fig. 3L, M). Attenuated expression of pro-tumorigenic genes (SLUG, vimentin, TWIST, ZEB1) was observed in R196Q (Fig. 3N). The transcriptional signature and all the functional properties indicated R196Q being a partial gain-of-function mutation which exhibits moderate resistance to cisplatin-induced apoptosis and fails to undergo the cell cycle arrest, further supporting the functional impairment of p53-mediated stress responses in this mutant.

p53 mutations exhibit a complex regulation of the PIK3CA gene that dictates sensitivity to cisplatin- alpelisib treatment

Using cBioPortal, we found a positive correlation between basal PIK3CA expression and p53 mutants-R282W, R175H, R273C, S241F, I195T, V157F, Y220C, H179R (compared to Wtp53). R273H, R248W, H193R, and C176Y mutants were negatively correlated with PIK3CA expression, and R248W mutant showed no association (Fig.S5F). Next, a small library of oncogenic mutant p53s (R175H, R282W, R273C, R273H, and R248W) and Wtp53 was created in p53-null SKOV3 cells (Fig. 4A). We also included the novel TP53 mutation-R196Q identified from our patient cohort. All mutants showed stabilization in untreated conditions with minimal changes after cisplatin treatment (Fig. 4B). Cisplatin induced a slight (2-4-fold) increase in p21 expression across mutants, except for SKOV3-R282W (22-fold) and SKOV3-Wtp53 (25-fold) (Fig. S4A).

The introduction of Wtp53 significantly enhanced the platinum sensitivity of SKOV3 cells (IC₅₀-3.4µM compared to 11.1µM) (Fig. 4C, D). In contrast, mutant-expressing cells showed cisplatin resistance (R248W-7µM, R273C-12.6µM, R273H-11.6µM, R196Q-8.9µM), with R282W (23.8µM) and R175H (15.4µM) being the most resistant. We also assessed the efficacy of cisplatin in cells harbouring endogenous Wtp53 and mutant p53. Wtp53 bearing A2780 cells (IC₅₀-2.9µM), exhibited comparable cisplatin sensitivity to SKOV3-Wtp53, while TOV112D-R175H (IC₅₀-6.6µM), showed higher platinum resistance (Fig. S7A) than both A2780-Wtp53 and SKOV3-Wtp53 cells. TOV112D EOC cells indigenously possess R175H mutation of p53 and are uniquely suited to model the specific patient population we characterized as responders, namely, those with oncogenic GOF p53 mutations and non-mutated yet transcriptionally elevated PIK3CA expression. The long-term survival assays revealed that highly resistant mutants (R282W, SKOV3-R175H, and TOV112D-R175H) formed colonies at higher concentrations (1.5µM), whereas moderately resistant mutants (R248W, R273C/H) formed colonies only at lower cisplatin concentrations (0.5-1.0µM). SKOV3-Wtp53 and A2780 colonies were only observed at 0.5µM cisplatin (Fig. 4D, FigS4B).

Compared to the basal level, mutants differentially modulated PIK3CA transcription after cisplatin treatment. Wtp53 and mutants R248W, R273C and R273H exhibited reduced PIK3CA expression (1.6- fold, 2.5-fold, 5.6-fold, 2.6-fold, respectively, whereas parental SKOV3 cells showed 1.2-fold induction. The cisplatin-resistant mutants R282W, R175H, and R196Q showed increased PIK3CA expression (2.1-fold, 1.8-fold, and 3.4-fold, respectively) after platinum treatment (Fig. 4E). Concurring with the transcript data, p110α and pAKT levels were elevated in SKOV3 cells, R282W, and R175H (Fig. 4F), while R248W and R273C/H showed slight reductions in p110α and a pronounced decrease in pAKT levels (2.5, 1.9, and 1.8-fold, respectively) (Fig. 4G). Total AKT level remained consistent across all cells, regardless of cisplatin treatment.

Low doses (IC₂₀-IC₃₀) of cisplatin-alpelisib combinatorial treatment exhibited a higher decrease in cell viability (~ 1.6-fold) in R282W, R175H, and R196Q mutants compared to parental SKOV3 cells (1.3-fold) (Fig. 4H) and enhanced PARP cleavage (Fig. 4J). Other mutants and Wtp53 showed no response to this combination (Fig. 4I), although Wtp53 and R273C (non-responder) showed PARP cleavage with cisplatin alone (Fig. 4K). Immunoblotting revealed marked downregulation of pS9-GSK3B, a critical AKT target, in all responders (Fig. 4L), suggesting that apoptosis, marked by PARP cleavage, may occur via the GSK3B pathway. Non-responders showed no significant alterations in pS9-GSK3B (Fig. 4M). No trend in Phospho-S6 kinase levels were seen in both Responders and Non-Responders (Fig. 4L, M).

Differential binding of mutant p53 and NFκB modulates PIK3CA activation after cisplatin treatment

Wtp53, when stabilized by platinum agents, represses PIK3CA by binding to its promoter region [19, 41, 42]. However, the functional attributes of mutant p53 proteins in this context remain poorly characterized. We employed chromatin immunoprecipitation assay to assess the binding efficiency of Wtp53 and selected mutants at four known p53 response elements within the PIK3CA promoter. Given that p53-mediated gene regulation is often concentration- and time-dependent, a cutoff of 1.5-fold was arbitrarily set, and any change below 1.5-fold was not considered.

Cisplatin drastically increased Wtp53 occupancy at all four sites (16-fold at S1, 28.5-fold at S2, 50-fold at S3, and 30-fold at S4) on the PIK3CA promoter (Fig. 5 B-E). Conversely, mutants associated with cisplatin-induced PIK3CA upregulation demonstrated reduced occupancy at sites S1, S2, S3, and S4 compared to untreated conditions. R282W showed a 3.6-, 3.2-, 2.7, and 3.7-fold decrease at S1, S2, S3 and S4 respectively. Similarly, the R175H mutant showed significantly diminished binding at S1 (27-fold), Site 2 (10-fold), S3 (21.0-fold), and S4 (13-fold). TOV112D also showed decreased promoter occupancy at 3 sites (3-fold, at S1 and S2 and 2.3-fold at S4), similar to SKOV3-R175H. R196Q demonstrated reduced promoter occupancy by around 4.7-fold at S1, 1.6-fold at S2, and 1.6-fold at S3, with no significant change observed at S4 (Fig. 5). Interestingly, for R273C, where cisplatin reduced PIK3CA expression, higher promoter occupancy was observed at S1(1.5-fold) S2(1.9-fold) and S3(2-fold) (Fig. 5B-D), indicating that this mutant might have retained partial wild-type function.

Fig. 5 — Differential binding of mutant p53 and NFκB modulates *PIK3CA* activation after cisplatin treatment(A) *PIK3CA* promoter schematic showing the four p53 binding sites on the *PIK3CA* promoter. (B-E) ChIP-qPCR data showing change in p53 occupancy on *PIK3CA* promoter at (B) Site 1 (C) Site 2 (D) Site 3 and (E) Site 4 post cisplatin treatment at their respective IC₅₀ concentrations. (F-G) ChIP-qPCR data showing changes in NFκB occupancy on *PIK3CA* promoter in TOV112D(R175H) and SKOV3-R273C cells at (F) Site 2 and (G) Site 4. (H-I) Data represented as mean ± SEM of n = 2 biological replicates. Statistical analysis for ChIP-qPCR data was done using unpaired TTest. (H-I) Representative images of NFκB and p53 localization by immunofluorescence in (H) TOV112D-R175H and (I) SKOV3-R273C (J) Quantification of colocalized NFκB and p53 puncta per nucleus of TOV112D and R273C cells, analysed using ImageJ-Fiji ComDet Plugin. 50 cells per condition was counted. Statistical analysis was performed by one-way ANOVA. (K) Immunoblot representing NFκB and p53 levels in nuclear, cytoplasmic and whole cell extracts in TOV112D and SKOV3-R273C cells. Lamin A/C was used as nuclear control and α-tubulin was used as cytoplasmic control. (L) Graph depicting *PIK3CA* transcript levels in wild type and NFκB knockdown and overexpressing TOV112D and SKOV3-R273C cells. (M) Graph showing percentage cell viability with alpelisib, cisplatin and combination treatment in parental, NFκB knockdown and NFκB overexpressing TOV112D and SKOV3-R273C cells. Data represented as mean ± SEM of n = 3 biological replicates, *p < 0.05, **p < 0.01, ***p < 0.001, n.s.- non-significant, as calculated by unpaired TTest (N-O) Immunoblot showing the protein levels of *PIK3CA*, PAKT, TAKT and cleaved PARP in parental, NFκB knockdown and NFκB overexpressing (N) TOV112D and (O)SKOV3-R273C cells with cisplatin, alpelisib and combination treatment

To understand the mechanism underlying cisplatin mediated increase in the PIK3CA transcription by the responder mutants, we performed promoter analysis wherein we identified NFκB (a known activator of PIK3CA) response elements, overlapping with p53 binding sites 2 and 4 (Fig. 5A). Intriguingly, while p53 occupancy decreased on sites 2 and 4, NFκB binding to the PIK3CA promoter at these two sites increased by 1.5 and 1.7-fold in TOV112D cells, respectively. In contrast, in the non-responder, SKOV3-R273C cells, NFκB binding significantly decreased by 3.9-fold and 1.3-fold, respectively with a concomitant increase in p53 occupancy (Fig. 5F, G).

Dual immunofluorescence assay also showed a 5-fold increase in the nuclear colocalization of NFκB with mutant p53 after cisplatin treatment in TOV112D cells, while colocalized puncta decreased by 1.9-fold in the non-responder, SKOV3-R273C cells (Fig. 5H-J). Nuclear colocalized puncta also increased by 1.9-fold in SKOV3-R175H, and decreased by 2.4-fold in another non-responder SKOV3-R248W cells. Such colocalization might imply concomitant occupancy of both the transcriptional regulators on the chromatin. No change was observed in Wtp53 expressing cells (Fig. S5B-E). Additionally, cellular fractionation revealed an increase in accumulation of NFκB in the nucleus in the TOV112D cells following cisplatin treatment, while the nuclear localization of NFκB decreased significantly in the SKOV3-R273C cells. Cytoplasmic levels of NFκB, however, remained unchanged (Fig. 5K). Further, we knocked NFκB down by shRNA mediated approach and then overexpressed the gene to validate our p53- NFκB interaction for PIK3CA regulation in TOV112D (responder) and SKOV3-R273C (non-responder) cells (Fig S5G). In parental TOV112D cells, cisplatin treatment upregulated PIK3CA transcript levels, and this upregulation was abrogated upon NFκB knockdown (TOV112D^KD), demonstrating that NFκB is required for cisplatin-induced PIK3CA expression in responder cells (Fig. 5L). Importantly, overexpression of NFκB in the knockdown background (TOV112D^OE) not only rescued but further amplified PIK3CA transcript levels upon cisplatin treatment, confirming that NFκB transcriptionally drives PIK3CA expression in these cells. TOV112D^KD cells failed to show response to alpelisib-cisplatin combination treatment and this sensitivity was restored upon overexpression of NFκB, as measured by the cytotoxicity assay (Fig. 5M) and increased PARP cleavage (Fig. 5N). In contrast, cisplatin induced reduction in PIK3CA expression did not change with NFκB knockdown and overexpression in SKOV3-R273C cells. These cells also did not show any differential response to combination therapy compared to monotherapy (Fig. 5M, O).

Altogether our data suggest that cisplatin promotes nuclear localization and binding of NFκB to the PIK3CA promoter by reducing promoter occupancy of specific mutant p53, leading to enhanced PIK3CA transcription and increased sensitivity to cisplatin- alpelisib combination.

Cisplatin-alpelisib treatment effectively reduced tumor growth than cisplatin in TOV112D-R175H tumor xenografts

Next to assess the in vivo efficacy of the combination therapy, mice harbouring platinum resistant TOV112D and A2780^LR tumors, were subjected to a short 8-day treatment regime followed by a 10-day treatment-free period (Fig. 6A). Platinum resistant TOV112D cells, inherently harbor p53-R175H mutation while A2780^LR endogenously harbors Wtp53. Interestingly, both combination (4.4-fold) and alpelisib monotherapy (3.2-fold), significantly reduced tumor volume compared to the vehicle control at the end of treatment for TOV112D xenografts. Tumour volume in the combination cohort was also 3.2-fold lower than the cisplatin cohort (Fig. 6.B, D). Cisplatin-alpelisib treatment exerted greater tumor growth reduction than alpelisib monotherapy; however, it did not meet statistical significance. In contrast, no significant change was observed among the cisplatin, alpelisib, and combination treatment groups in the A2780^LR tumour-bearing mice (Fig. 6C, E).

Fig. 6 — Cisplatin-alpelisib combinatorial treatment effectively reduced tumor growth than cisplatin in TOV-112D-R175H tumor xenografts: (A) Schematic representation of the treatment regimen for assessing cisplatin-alpelisib efficacy for TOV112D and A2780^LR, created using Biorender (ON29VA9K3N). (B-C) Graph representing tumor volume of vehicle, cisplatin, alpelisib, and combination-treated mice bearing (B) TOV112D on Day 11(pre-treatment), Day 18, and 21(treatment period), and Day 31(Post-Holiday Period) and (C) A2780^LR on Day 14(pre-treatment), Day 20, and 23(treatment period), and Day 33(Post-Holiday Period). Cisplatin treatment was given intraperitoneally on Days 14 and 19 for TOV112D and Days 15 and 21 for A2780^LR. Alpelisib was given orally daily for seven days (days 15–21 for TOV112D and Days 16–22 for A2780^LR). (D-E) Graph representing fold change in tumor volume at (D) Days 18, 21, and 31, with respect to Day 11(pre-treatment) for TOV112D tumour-bearing mice and (E) Days 20, 23, and 33, with respect to Day 14(pre-treatment) for A2780^LR tumour-bearing mice. The tables in B-E show P-values at the end of the treatment period and at the end of the holiday period across cohorts. (F-G) Graph representing the change in body weight in mice during the study period for (F) TOV112D and (G)A2780^LR. Statistical analysis was performed using two-way ANOVA followed by Sidak’s multiple comparisons test. Tumor volumes were measured using digital calipers and calculated using the formula: V=(length×width^2)/2. Data represented as mean ± SEM, p-values are tabulated next to the graphs, p ≤ 0.05 was considered significant

During the holiday period, tumor growth in the TOV112D mice persisted in all groups, with the alpelisib cohort showing the highest increase (3.8-fold), suggesting a potential growth advantage upon discontinuation of the drug (Fig. 6B, D). Tumor volumes also increased significantly in the vehicle and cisplatin groups but remained non-significant in the combination group (Fig. 6B, D). Histopathological evaluation of Haematoxylin and Eosin-stained tumor sections (Fig.S6 C-F), revealed significant differences in necrosis levels across the cohorts. The combination-treated tumors showed maximum necrosis (~ 74%) than the other three groups (40% for vehicle control, 48.8% for cisplatin, and 39% for alpelisib) at the end of the holiday period (Fig.S6A). In fact, only ~ 30% necrosis was observed in the three combination-treated mice sacrificed immediately after the treatment phase (Fig. S6B). In-vitro Annexin-PI assay with TOV112D cells showed a significantly higher necrotic population in the combination cohort after a 48-hour recovery period post-24-hour treatment of cisplatin-alpelisib (Fig.S6H). Necrosis in cisplatin, alpelisib, and cisplatin-alpelisib-treated cells after 24 h of treatment remained unchanged (Fig.S6G). In A2780^LR xenografts, a slight increase in tumor volumes for cisplatin, alpelisib, and cisplatin-alpelisib, and a significant increase in the untreated groups were observed. No difference in the tumor growth for each group compared between the end of treatment and the end of the holiday period was noticed.

Overall body weight in TOV112D model were significantly increased in the control and cisplatin-treated groups, correlating with their tumor volume growth, but not in any of the mice of alpelisib or combination cohorts (Fig. 6F, G). In contrast, no noticeable change in body weight was observed among all four groups for A2780^LR xenografted mice till the end of the holiday period. Although some mice in the alpelisib and combination groups showed hair loss, no diarrhoea or nephrotoxicity was observed (Fig.S6I). These findings suggest that the combination of cisplatin and alpelisib was well-tolerated at the administered doses. The enhanced anti-tumor efficacy, reflected by increased necrosis and reduced tumor growth in TOV112D xenografts, persisted beyond the treatment period, suggesting the potential long-term benefits of this combination therapy.

Cisplatin-Alpelisib treatment decreases ascites burden in patient-derived xenograft model

We established a malignant-ascites generating PDX model from treatment-naïve patient-derived primary cells(P16). NGS revealed a similar mutational landscape between the patient-derived primary tumour cells and PDX-derived tumor cells that retained the same frameshift mutation in BRCA1 and a R175H mutation in TP53 (Fig. 7A). Further, these cells showed PIK3CA amplification (relative copy number > 2) compared to all three reference genes (Fig. 7B).

Fig. 7 — Cisplatin-alpelisib combination shows increased cytotoxicity in R175H expressing PDX Spheroids (A) Oncoprint plot showing somatic pathogenic mutations in P16 and PDX from P16. (B) Graph depicting the *PIK3CA* copy number in P16 and PDX-P16 cells relative to either *GAPDH*, *ACTB*, or *COL8A1* from IOSE364 cells (C) Graph depicting the body weight across control, cisplatin, alpelisib, and combination-treated PDX-P16 mice. (D) Graph depicting fold change in body weight across control, cisplatin, alpelisib, and combination-treated PDX-P16 mice. Statistical analysis was performed using one-way ANOVA, p-values are tabulated under the graphs and p ≤ 0.05 was considered significant. (E)Images showing ascites collected (left) and spheroids isolated (right) from the ascites of PDX-P16 mice across groups (F) Bar graph showing volume of ascites collected from PDX-P16 mice in different groups (G-H) Cell viability assay showing the responses to respective IC₂₀ concentrations of cisplatin, alpelisib, alone and in combination for (G) PDX-P16 and (H) TOV112D. Data represented as mean ± SEM of n = 3 replicates, Statistical analysis was performed using students unpaired TTest, *p < 0.05, **p < 0.01, ***p < 0.001, (I-J) Immunoblot depicting changes in the levels of p110α, p-AKT, p53 and cleaved PARP by combination treatment (at IC₂₀ concentrations) of cisplatin and alpelisib in (I) PDX-P16 and (J) TOV112D. (K-L) Immunoblot depicting changes in the levels of GSK3β and S6K by combination treatment (at IC₂₀ concentrations) of cisplatin and alpelisib in (K) PDX-P16 and (L) TOV112D. (M) Quantification of spheroid images at untreated and IC₂₀ concentrations of cisplatin, alpelisib, and combination treatment at 24–72 h post-treatment. Data represented as mean ± SEM or n = 3 replicates, Statistical analysis was performed using students unpaired TTest, p-values are denoted above the graph. p ≤ 0.05 was considered significant. (N) Representative Bright field images of Untreated and cisplatin, alpelisib, and cisplatin-alpelisib treated Spheroids at 24–72 h post-treatment

In vivo efficacy of cisplatin-alpelisib was assessed in this PDX mouse model following the treatment regimen as described earlier. The combination therapy was most effective with a 29.4% reduction in body weight compared to vehicle controls, followed by cisplatin alone at 17.1% reduction. Alpelisib monotherapy showed minimal effect, with only a 4.4% decrease in body weight (Fig. 7C, D). Combination treatment also showed 14.8% lower weight compared to cisplatin alone, highlighting the efficacy of this combination over monotherapies. The amount of malignant ascites present was significantly lower in cisplatin and cisplatin-alpelisib-treated mice than in untreated and alpelisib-treated mice (Fig. 7E, F, S7C). The number of tumor spheroids in ascites was significantly reduced in the cisplatin-treated group and was barely present in the combinatorial treatment group compared to the untreated and alpelisib-treated mice (Fig. 7E). Metastatic tumours were observed in the peritoneum, intestines and spleen of the control tumours but not observed in any of the treatment groups (data not shown).

Unfortunately, we were not able to assess the effect of cisplatin, alpelisib, and the combination due to the rapid senescence of the patient-derived (P16) tumor cells in culture. However, the PDX-derived (P16) tumor cells showed similar cisplatin sensitivity (IC₅₀-5 μm) to TOV112D (IC₅₀-6.5 μm) cells that endogenously harbor the R175H mutation (Fig. S7A). For both, the IC₅₀ value of alpelisib (between 1 and 10µM) was not reached (Fig. S7B). Combination treatment with cisplatin and alpelisib further reduced viability, with a 1.2-fold and 1.5-fold decrease in PDX and TOV112D cells, respectively, compared to monotherapies (Fig. 7G, H). Immunoblot analysis after drug treatment showed that both PDX and TOV112D resulted in higher p110α expression and pAkt levels with cisplatin treatment and significant PARP cleavage with combination treatment (Fig. 7I, J). Additionally, consistent with other responders, downregulation of pS9-GSK3β was observed in both p53 R175H-expressing PDX and TOV112D cells (Fig. 7K, L).

To assess the sensitivity of the spheroids to this combinatorial treatment, a similar number of tumor spheroids from the untreated mice were treated with IC₂₀ concentrations of cisplatin and alpelisib, either individually or in combination, for 72 h. While a slight, statistically non-significant increase in cell death was observed at 24 h, prolonged drug exposure resulted in a substantial reduction in spheroid viability. Cisplatin treatment decreased cell viability at 48 and 72 h. However, combination treatment induced a significantly greater decrease in viability than cisplatin, with a 2.5-fold and 2.7-fold increase in cell death observed at 48 and 72 h, respectively. Only a slight reduction in the spheroid count was observed with alpelisib treatment at 48 and 72 h (Fig. 7M, N).

Discussion

The elegant era of omics for cancer biology has facilitated the development of several targeted therapies, improving the treatment outcomes and extending progression-free survival. Yet, certain cancers and cancer subtypes, including High-grade serous ovarian cancer (HGSOC), remain challenging to treat owing to their heterogeneity, late detection, and acquired chemoresistance. Though PARP inhibitor treatment immensely benefits around 50% of HGSOC cases harboring BRCA1/2 mutations or HRD, they are now predominantly used as upfront maintenance therapy in HRD/BRCA-mutated patients following first-line chemotherapy response. Thus, newer therapeutic interventions are needed for the platinum resistant patients [43]. Herein, we evaluated the combinatorial effect of cisplatin and alpelisib, the sole clinically approved PIK3CA inhibitor, in a small cohort of treatment-naïve and recurrent HGSOC patient-derived tumor cells, cell lines, tumor xenografts, and PDX mouse model. Our data revealed that cisplatin-alpelisib combination exerts superior response than cisplatin or alpelisib in 33% of the malignant ascites derived primary tumor cells, majorly harboring PIK3CA gene amplification (~ 80% of the responders) but no pathogenic mutation, and somatic BRCA2 mutations (~ 75% of the responders). We also showed that specific p53 mutations (R196Q, R282W and R175H) could enhance PIK3CA transcription through a co-operative and reciprocal promoter binding activity with NFκB, thereby, generating a significant response towards alpelisib-cisplatin treatment in mutant or Wt PIK3CA background. Alpelisib-cisplatin combination effectively reduced the ascites burden in a PDX mouse model harboring p53-R175H as well as arrested initial growth of platinum-resistant TOV112D tumors (p53-R175H) with continued growth inhibition for 10 days after completion of treatment. This is probably the first report on alpelisib-induced sensitization to cisplatin in HGSOC, suggesting that upregulated PIK3CA signaling, either through transcriptional regulation or amplifications, can predict response to PIK3CA inhibitors (Fig. 8).

Fig. 8 — Differential regulation of *PIK3CA* Promoter by NFκB and mutant-p53 drives sensitivity to combination of cisplatin and alpelisib: (Left) In case of responder mutants, cisplatin treatment promotes nuclear localization and binding of NFκB to the *PIK3CA* promoter by reducing promoter occupancy of specific mutant p53, leading to enhanced *PIK3CA* transcription and increased apoptosis and sensitivity to cisplatin- alpelisib combination. (Right) In case of the non-responders, cisplatin treatment increases mutant p53 binding to the *PIK3CA* promoter, reducing NFκB binding and leading to a suppression in *PIK3CA* transcription, thereby leading to resistance to cisplatin and alpelisib combination therapy. Created using BioRender (DY29VA7WZY)

Genome analysis of Indian HGSOC patients is still in its infancy, and our study though in a small cohort is probably the first such in-depth analysis for both germline and somatic mutations in treatment-naïve and relapsed patients. Previous studies of ~ 140 Indian patients with HGSOC, only investigated germline mutations using a 14-gene panel [44, 45]. Though our findings align with TCGA reports regarding HR pathway alterations and germline BRCA1/2 mutations, somatic BRCA2 mutations were more prevalent (42%), warranting validation in a larger cohort. A pathogenic germline mutation in PTCH1, mutations in AKT1 (21%), and ATRX (29%), but no pathogenic mutation in PIK3CA or PTEN were found.

Around 34% of HGSOC patients harbor an activated PIK3CA-AKT signaling majorly contributed by copy number amplification (~ 27%) which are mutually exclusive with BRCA1/2 or HRD and are potential targets for PIK3CA inhibitors [10]. Alpelisib is an orally bioavailable, a-specific phosphatidylinositol-3-kinase (PI3K) inhibitor that showed good efficacy in inhibiting the growth of PI3Kα driven tumors in animal xenograft models [16, 46]. This inhibitor is currently approved for patients harboring PIK3CA activating mutations and already in use for PIK3CA mutant, HR+ HER2- Breast Cancer in combination with Fulvestrant. Very few studies have explored the efficacy of alpelisib, against various genomic alterations of PIK3CA gene. Fritsch et al., showed that PIK3CA mutation served as the foremost positive predictor of alpelisib (NVP-BYL719) sensitivity with additional positive and negative associations such as PIK3CA amplification and PTEN mutation. PIK3CA amplifications [16] showed a reduced likelihood of remission and platinum sensitivity in patients treated with taxane-platinum chemotherapy [47]. Several studies have implicated PIK3CA pathway activation as a driver of platinum resistance across gynaecological cancers either through PIK3CA copy number amplifications [47], transcriptional upregulation of PIK3CA combined with PTEN loss [48]and also through activating mutations in PIK3CA [49, 50]. Collectively, these findings establish PIK3CA/AKT pathway activation as a recurrent mechanism of platinum resistance, yet none of these studies have examined this in the context of any transcriptional regulators, specifically p53 mutant variants, representing a key gap that our present study addressed.

Alpelisib, as monotherapy, due to its cytostatic nature, only arrests the growth and proliferation of cells but is not effective in inducing cell death. Combining PI3K inhibition with 5-fluorouracil (5-FU) or Paclitaxel enhanced anticancer effects in cancer cell lines [51]. Though EPIK-O, an open-label, phase III trial consisting of 358 platinum refractory/platinum resistant HGSOC patients with Wt BRCA did not show improved progression free survival (PFS) or overall survival (OS) in alpelisib-olaparib arm compared to standard chemotherapy (paclitaxel or doxorubicin) arm, 33% of PIK3CA altered tumors showed higher response to alpelisib-olaparib compared to the 16% to standard therapy [17]. PI3K, through its downstream effector, AKT, induces DNA repair by activating DNA-PKcs [56] and RAD51 [57]. Thus, we reasoned that combining cisplatin, a DNA-damaging agent, with alpelisib may effectively block DNA repair and further sensitize cells to cisplatin-induced apoptosis. Indeed, among the 5/15 (33%) responders in our cohort, four harbored PIK3CA gene amplification (~ 26%) which is at par with TCGA data[10]. Validation of these findings in a larger cohort will help to identify the percentage of responders more accurately.

Our TCGA analysis revealed that around ~ 6.8% of HGSOC patients possess higher PIK3CA transcript levels without any genomic alterations, suggesting an alternative mode of activation of this oncogenic pathway. Among the multiple positive and negative regulators of PIK3CA expression, Wtp53 acts as a repressor through enhanced promoter occupancy after platinum treatment [19, 41]. However, the active role of mutants of p53 in modulating PIK3CA expression is largely unknown. Identification of a novel p53 mutation (R196Q) in our responder cohort and association of certain oncogenic p53 mutations with heightened PIK3CA transcription in TCGA datasets suggest for a critical function of p53 mutants in PIK3CA biology. R196Q is annotated as a rare, partially functional TP53 mutation in the IARC/TP53 database, occurring in approximately 0.04% of all cancers. Although previously reported in lymphoma, bladder cancer, and breast cancer, it has not been previously reported in ovarian cancer. The low variant allelic frequency observed in our patient sample (3.4%) is indicative of sub-clonal population, suggesting that R196Q-harboring cells constitute a minor but functionally distinct population within the tumour. These sub clonal populations might carry potential therapeutic implications, as partially functional variants and may contribute to acquirement of resistance under the selective pressure of platinum-based chemotherapy. Importantly, our data demonstrates that R196Q-expressing cells retain sensitivity to the cisplatin-alpelisib combination, suggesting that the targeted suppression of PI3K signalling in conjunction with platinum-based therapy may represent an effective strategy for treating platinum resistant tumours harbouring this mutation.

R196Q mutant expressing cells exhibited a distinct transcriptional signature, enhanced migration but a non-invasive and non-tumorigenic phenotype indicative of possessing partial gain-of-function when compared to Wtp53. Overexpression of R196Q, R175H and R282W mutations in a p53-null SKOV3 cells led to increased PIK3CA transcription by cisplatin and greater sensitivity to cisplatin-alpelisib combination therapy compared to parental SKOV3 cells, which harbor an activating PIK3CA mutation. Such incremental response towards alpelisib-cisplatin by certain p53 mutants but not by the others (R248W, R273C, R273H as well as Wt p53) in a backdrop of activating mutant PIK3CA suggests for a complex regulation of PIK3CA promoter. TOV112D cells indigenously harboring p53-R175H and Wt PIK3CA, also showed a similar trend. PIK3CA is transcriptionally activated by SOX2, SOX4 [58, 59], DDX55, NFKB1[60, 61], YB1 [62], and CUX1 [63] and, repressed by Wtp53. Previous studies, including our own, have shown that the PIK3CA promoter contains multiple p53 binding sites and context-dependent (temperature or drug-induced) binding of stabilized p53 repressed promoter activity and downregulated PIK3CA expression [41, 19] . We have also showed that cisplatin induced nuclear NFκB could bind in this promoter region and activates PIK3CA in absence of binding of Wtp53 in platinum resistant EOC cells [52]. Though arising from a single protein, p53 mutants are now recognized as a family of proteins with overlapping and distinct transcriptional signatures, significantly influencing cancer initiation, progression, drug sensitivity, and other characteristics [64, 65]. Our current ChIP data suggest a similar differential regulatory role for individual p53 mutants on transcriptional regulation of PIK3CA. We also observed a co-operative reciprocal activity between NFκB and R175H mutant where lesser degree of promoter occupancy by mutant p53 allowed NFκB to bind and transcriptionally upregulating PIK3CA gene. This mechanism is strongly supported by previous reports demonstrating that p53-R175H bearing cells exhibit increased nuclear localization of NFκB [66] and promote a pro-inflammatory phenotype through enhanced NFκB activation [67]. When NFκB expression was depleted, abrogation in transcriptional surge in PIK3CA was only observed for R175H but not for R273C mutant. Overexpression of NFκB in this knockdown background rescued PIK3CA upregulation only in R175H mutant expressing cells. The alpelisib-cisplatin sensitivity also got modulated by alteration in NFκB level in R175H but not in R273C. These mutant specific responders primed the cisplatin-alpelisib-treated cells towards apoptosis via diminished phospho-AKT and activation of GSK3B through reduced inhibitory serine9 phosphorylation [53, 54]. In non-responders, GSK3B remained inhibited, indicated by persistent S9 phosphorylation leading to resistance towards combination treatment. Understanding the complex interplay between NFκB and mutant-p53 and their co-regulation of the PIK3CA promoter through ChIP-seq and immunoprecipitation is still under investigation.

This mutant p53 specific differential regulation of PIK3CA-AKT signaling produced significant treatment efficacy of cisplatin-alpelisib in pre-clinical tumor xenograft (TOV112D) and PDX models, both carrying R175H mutation. While alpelisib alone and in combination significantly arrested TOV112D tumor growth compared to cisplatin and control, the effect was sustained only for the combinatorial regime during the treatment-free period. Such a transient anti-proliferative effect possibly resulted due to the G0-G1 phase arrest by the inhibitor, which was overridden by cessation of alpelisib treatment [51]. Tumor necrosis was ~ 2-fold higher in the combination group, suggesting a long-term therapeutic benefit of cisplatin-alpelisib. Similar effects by alpelisib or cisplatin-alpelisib were not evident in A2780^LR tumor xenografts, which harbor a Wtp53. The most clinically relevant effect was observed in the PDX model where combination treatment effectively reduced the body weight and malignant ascites formation compared to monotherapies. Even tumor spheroids, a major component of malignant ascites which are highly resistant to chemotherapy, due to poor drug penetrance, enhanced autophagy, and stemness showed significant mortality in vitro by alpelisib-cisplatin than monotherapies alone. This is evident by the reduction in the number and size of spheroids in the combination treatment group at 48- and 72-hours post-treatment, indicating that this combination may help in abating the spheroid burden in HGSOC patients. The absence of long-term survival data with the combination treatment to assess the tumour growth kinetics and associated drug toxicities, if any, will be tested in future.

Transcriptional activation of a gene affects therapy by modulating downstream signalling which often falls short to be considered over the recognizable genetic alterations (mutation or copy number alteration). TCGA data collected from c-bioportal (310 cases) indicated that ~ 6.8% of HGSOC patients harbour elevated PIK3CA mRNA without any underlying genomic alteration. This constitutes a subset of patients that would be missed by the current genomic screening approaches focusing only on mutations or copy number alterations in patients. Thus, it is important to account for a clinically actionable biomarker which would predict the transcriptional activation of PIK3CA gene. In our study, we have identified a few prominent p53 mutations which could be used as the predictive biomarkers for PIK3CA transcriptional activation. Till date, mutation specific antibodies for p53 gene are not commercially available except for R175H mutation which currently limit this strategy. We believe that availability of more mutation specific p53 antibodies in future will open up a new screening approach to select patients for PIK3CA based therapy.

This is the first report demonstrating the efficacy of alpelisib with cisplatin in a copy number amplified and mRNA-high but non-mutated PIK3CA background, highlighting the role of mutant p53 in modulating this pathway. This combination therapy represents a promising novel therapeutic strategy for HGSOC patients, demonstrating significant antitumor efficacy and reduction of ascites burden in preclinical in vivo mouse models and increased cytotoxicity in patient-derived cells and tumour spheroids. Validation in a larger cohort in preclinical models and clinical samples is warranted to understand long term potential of this strategy.

Supplementary Information

Supplementary Material 1.^{(3.2MB, docx)}

Supplementary Material 2.^{(2.2MB, pdf)}

Acknowledgements

This study was funded by the Department of Biotechnology (Grant No: BT/PR27126/MED/30/1943/2017) and DAE-ACTREC (Grant No: 1/3(7)/2020/TMC/R&D-II/8823 Dt.30.07.2021). M.M, S.C, E.S, and D.J acknowledge ACTREC for Fellowship. We also acknowledge the FACS, Digital Imaging, Molecular Imaging and Animal House Facility at ACTREC. We thank Dr. Prasanna Venkatraman for the pEBB-HA RelA plasmid. We thank all the patients from whom ascites were collected for this study.

Abbreviations

HGSOC: High-Grade Serous Ovarian Cancer
ChIP: Chromatin Immunoprecipitation
PFS: Progression-free survival
TCGA: The Cancer Genome Atlas
HRR: Homologous Recombination Repair
PIK3CA: Phosphatidylinositol-4,5-Bisphosphate 3-Kinase Catalytic Subunit Alpha
Wtp53: Wild-type p53
OS: Overall Survival
HRD: Homologous Recombination Deficient

Authors’ contributions

MM : Investigation, Writing - Original Draft, Visualization. SC: Investigation. ES : Formal analysis, Visualization. DJ: Investigation AD: Investigation. AS: Resources. PC: Formal analysis. AM: Methodology, Resources. SM: Formal analysis. JG: Methodology, Resources, SG: Methodology, Resources, and PR : Conceptualization, Methodology, Writing - Review & Editing, Supervision, Funding acquisition. All the authors have read and approved the final version of the manuscript.

Data availability

The sequencing raw data can be accessed from ENA/SRA using accession number PRJEB71525.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Ethics Committee of ACTREC-TMC (IEC#900513), in compliance to the guidelines provided by the Declaration of Helsinki. Written informed consent was taken from all the patients from whom the samples were collected in this study.

All the animal studies were performed per the approval of the Institutional Animal Ethics Committee of ACTREC (IAEC#26/2020). The animals used in the study were provided by the Animal House Facility of ACTREC-TMC.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1.Borley J, et al. Does tumour biology determine surgical success in the treatment of epithelial ovarian cancer? A systematic literature review. Br J Cancer. 2012;107(7):1069–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. Cancer J Clin. 2020;70(1):7–30. [DOI] [PubMed] [Google Scholar]
3.Bast RC Jr., Hennessy B, Mills GB. The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer. 2009;9(6):415–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Herzog TJ. The current treatment of recurrent ovarian cancer. Curr Oncol Rep. 2006;8:448–54. [DOI] [PubMed] [Google Scholar]
5.Stuart GC et al. 2010 Gynecologic Cancer InterGroup (GCIG) consensus statement on clinical trials in ovarian cancer: report from the Fourth Ovarian Cancer Consensus Conference. Int J Gynecol Cancer. 2011;21(4):750-5. [DOI] [PubMed] [Google Scholar]
6.Mehrotra M, et al. Drug-Resistant Epithelial Ovarian Cancer: Current and Future Perspectives. Adv Exp Med Biol. 2024;1452:65–96. [DOI] [PubMed] [Google Scholar]
7.Dilawari A, et al. FDA Approval Summary: Mirvetuximab Soravtansine-Gynx for FRα-Positive, Platinum-Resistant Ovarian Cancer. Clin Cancer Res. 2023;29(19):3835–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Gupta S, et al. Maintenance therapy for recurrent epithelial ovarian cancer: current therapies and future perspectives – a review. J Ovarian Res. 2019;12(1):103. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Joris S, et al. Efficacy of olaparib in advanced cancers with germline or somatic mutations in BRCA1, BRCA2, CHEK2 and ATM, a Belgian Precision tumor-agnostic phase II study. ESMO Open. 2023;8(6):102041. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Bell D, et al. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474(7353):609–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Bertelsen BI, et al. Molecular analysis of the PI3K-AKT pathway in uterine cervical neoplasia: Frequent PIK3CA amplification and AKT phosphorylation. Int J Cancer. 2006;118(8):1877–83. [DOI] [PubMed] [Google Scholar]
12.Hahne JC, et al. Downregulation of AKT reverses platinum resistance of human ovarian cancers in vitro. Oncol Rep. 2012;28(6):2023–8. [DOI] [PubMed] [Google Scholar]
13.Ohta T, et al. Inhibition of Phosphatidylinositol 3-Kinase Increases Efficacy of Cisplatin in in Vivo Ovarian Cancer Models. Endocrinology. 2006;147(4):1761–9. [DOI] [PubMed] [Google Scholar]
14.Furman RR, et al. Idelalisib and Rituximab in Relapsed Chronic Lymphocytic Leukemia. N Engl J Med. 2014;370(11):997–1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.André F, et al. Alpelisib for PIK3CA-Mutated, Hormone Receptor–Positive Advanced Breast Cancer. N Engl J Med. 2019;380(20):1929–40. [DOI] [PubMed] [Google Scholar]
16.Fritsch C, et al. Characterization of the novel and specific PI3Kα inhibitor NVP-BYL719 and development of the patient stratification strategy for clinical trials. Mol Cancer Ther. 2014;13(5):1117–29. [DOI] [PubMed] [Google Scholar]
17.Konstantinopoulos PA, et al. Primary Analysis of EPIK-O/ENGOT-ov61: Alpelisib Plus Olaparib Versus Chemotherapy in Platinum-Resistant or Platinum-Refractory High-Grade Serous Ovarian Cancer Without BRCA Mutation. J Clin Oncol. 2025;43(26):2908–17. [DOI] [PubMed] [Google Scholar]
18.Shenoy PS, et al. Developing Clinically Relevant Acquired Chemoresistance Models in Epithelial Ovarian Cancer Cell Lines. Bio Protoc. 2022;12(3):e4310. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Gaikwad SM, et al. Non-invasive imaging of phosphoinositide-3-kinase-catalytic-subunit-alpha (PIK3CA) promoter modulation in small animal models. PLoS ONE. 2013;8(2):e55971. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Phadte P, et al. Autophagy-mediated ID1 turnover dictates chemo-resistant fate in ovarian cancer stem cells. J Exp Clin Cancer Res. 2024;43(1):222. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Andrews S. FastQC: a quality control tool for high throughput sequence data. 2010. Available online at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc.
22.Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Li H, Durbin R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics. 2009;25(14):1754–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.McKenna A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Van der Auwera GA, et al. From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline. Curr Protoc Bioinf. 2013;43(1110):p11101–111033. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Liu X, et al. dbNSFP v4: a comprehensive database of transcript-specific functional predictions and annotations for human nonsynonymous and splice-site SNVs. Genome Med. 2020;12(1):103. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Zuguang Gu, Roland Eils, Matthias Schlesner. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32(18):2847–9. [DOI] [PubMed]
28.Li H, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25(16):2078–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Koboldt DC, et al. VarScan: variant detection in massively parallel sequencing of individual and pooled samples. Bioinformatics. 2009;25(17):2283–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010;38(16):e164. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Gaikwad SM, et al. Differential activation of NF-κB signaling is associated with platinum and taxane resistance in MyD88 deficient epithelial ovarian cancer cells. Int J Biochem Cell Biol. 2015;61:90–102. [DOI] [PubMed] [Google Scholar]
32.Messeguer X, et al. PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics. 2002;18(2):333–4. [DOI] [PubMed] [Google Scholar]
33.Dhadve AC et al. Decoding molecular interplay between RUNX1 and FOXO3a underlying the pulsatile IGF1R expression during acquirement of chemoresistance. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease. 2020;1866(6):165754. [DOI] [PubMed]
34.Davila JI, et al. Frequent POLE-driven hypermutation in ovarian endometrioid cancer revealed by mutational signatures in RNA sequencing. BMC Med Genomics. 2021;14(1):165. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Briggs S, Tomlinson I. Germline and somatic polymerase ε and δ mutations define a new class of hypermutated colorectal and endometrial cancers. J Pathol. 2013;230(2):148–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Hino H, et al. Clinicopathological and mutational analyses of colorectal cancer with mutations in the POLE gene. Cancer Med. 2019;8(10):4587–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Voropaeva EN, et al. Clinical aspects of TP53 gene inactivation in diffuse large B-cell lymphoma. BMC Med Genomics. 2019;12(Suppl 2):35. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Kim HS, et al. Direct measurement of engineered cancer mutations and their transcriptional phenotypes in single cells. Nat Biotechnol. 2024;42(8):1254–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Holstege H, et al. High Incidence of Protein-Truncating TP53 Mutations in BRCA1-Related Breast Cancer. Cancer Res. 2009;69(8):3625–33. [DOI] [PubMed] [Google Scholar]
40.Leroy B, Anderson M, Soussi T. TP53 Mutations in Human Cancer: Database Reassessment and Prospects for the Next Decade. Hum Mutat. 2014;35(6):672–88. [DOI] [PubMed] [Google Scholar]
41.Astanehe A, et al. Mechanisms underlying p53 regulation of PIK3CA transcription in ovarian surface epithelium and in ovarian cancer. J Cell Sci. 2008;121(Pt 5):664–74. [DOI] [PubMed] [Google Scholar]
42.Singh B, et al. 53 regulates cell survival by inhibiting PIK3CA in squamous cell carcinomas. Genes Dev. 2002;16(8):984–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Tew WP, et al. PARP Inhibitors in the Management of Ovarian Cancer: ASCO Guideline. J Clin Oncol. 2020;38(30):3468–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Kadri MSN, et al. Mutational Landscape for Indian Hereditary Breast and Ovarian Cancer Cohort Suggests Need for Identifying Population Specific Genes and Biomarkers for Screening. Front Oncol. 2020;10:568786. [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Mannan AU, et al. Detection of high frequency of mutations in a breast and/or ovarian cancer cohort: implications of embracing a multi-gene panel in molecular diagnosis in India. J Hum Genet. 2016;61(6):515–22. [DOI] [PubMed] [Google Scholar]
46.Furet P, et al. Discovery of NVP-BYL719 a potent and selective phosphatidylinositol-3 kinase alpha inhibitor selected for clinical evaluation. Bioorg Med Chem Lett. 2013;23(13):3741–8. [DOI] [PubMed] [Google Scholar]
47.Kolasa IK, et al. PIK3CA amplification associates with resistance to chemotherapy in ovarian cancer patients. Cancer Biol Ther. 2009;8(1):21–6. [DOI] [PubMed] [Google Scholar]
48.Lee S, et al. Activation of PI3K/Akt pathway by PTEN reduction and PIK3CA mRNA amplification contributes to cisplatin resistance in an ovarian cancer cell line. Gynecol Oncol. 2005;97(1):26–34. [DOI] [PubMed] [Google Scholar]
49.McIntyre JB, et al. PIK3CA mutational status and overall survival in patients with cervical cancer treated with radical chemoradiotherapy. Gynecol Oncol. 2013;128(3):409–14. [DOI] [PubMed] [Google Scholar]
50.Arjumand W, et al. Phosphatidyl inositol-3 kinase (PIK3CA) E545K mutation confers cisplatin resistance and a migratory phenotype in cervical cancer cells. Oncotarget. 2016;7(50):82424–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Kim KJ, et al. PI3K-targeting strategy using alpelisib to enhance the antitumor effect of paclitaxel in human gastric cancer. Sci Rep. 2020;10(1):12308. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Thakur B, Ray P. Cisplatin triggers cancer stem cell enrichment in platinum-resistant cells through NF-κB-TNFα-PIK3CA loop. J Experimental Clin Cancer Res. 2017;36(1):164. [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Jacobs KM, et al. GSK-3β: A Bifunctional Role in Cell Death Pathways. Int J Cell Biology. 2012;2012(1):930710. [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Beurel E, Jope RS. The paradoxical pro- and anti-apoptotic actions of GSK3 in the intrinsic and extrinsic apoptosis signaling pathways. Prog Neurobiol. 2006;79(4):173–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Gunaratna, RT, et al. Dynamic role of the codon 72 p53 single-nucleotide polymorphism in mammary tumorigenesis in a humanized mouse model. Oncogene. 2019;38(18):3535–50. 10.1038/s41388-018-0630-4. [DOI] [PMC free article] [PubMed]
56.Toulany M, et al. Akt promotes post-irradiation survival of human tumor cells through initiation, progression, and termination of DNA-PKcs-dependent DNA double-strand break repair. Mol Cancer Res. 2012;10(7):945–57. 10.1158/1541-7786.MCR-11-0592. [DOI] [PubMed]
57.Mueck K, et al. Akt1 stimulates homologous recombination repair of DNA double-strand breaks in a Rad51-dependent manner. Int J Mol Sci. 2017;18(11). [DOI] [PMC free article] [PubMed]
58.Tang J, et al. SOX2 recruits KLF4 to regulate nasopharyngeal carcinoma proliferation via PI3K/AKT signaling. Oncogenesis. 2018;7(8):61. [DOI] [PMC free article] [PubMed]
59.Mehta GA, et al. SOX4 and SMARCA4 cooperatively regulate PI3k signaling through transcriptional activation of TGFBR2. npj Breast Cancer. 2021;7(1):40. [DOI] [PMC free article] [PubMed]
60.Yu B, et al. DDX55 promotes hepatocellular carcinoma progression by interacting with BRD4 and participating in exosome-mediated cell-cell communication. Cancer Sci. 2022;113(9):3002–17. [DOI] [PMC free article] [PubMed]
61.Yang N, et al. Transcriptional regulation of PIK3CA oncogene by NF-κB in ovarian cancer microenvironment. PLOS ONE. 2008;3(3):e1758. [DOI] [PMC free article] [PubMed]
62.Astanehe A, et al. The transcriptional induction of PIK3CA in tumor cells is dependent on the oncoprotein Y-box binding protein-1. Oncogene. 2009;28(25):2406–18. [DOI] [PubMed]
63.Wang, Z., et al., PIK3CA Is Regulated by CUX1, Promotes Cell Growth and Metastasis in Bladder Cancer via Activating Epithelial-Mesenchymal Transition. Front Oncol. 2020;10. [DOI] [PMC free article] [PubMed]
64.Kennedy MC, Lowe SW. Mutant p53: it’s not all one and the same. Cell Death Differ. 2022;29(5):983–7. [DOI] [PMC free article] [PubMed]
65.Walerych D, Lisek K, Del Sal G. Mutant p53: One, No One, and One Hundred Thousand. Front Oncol. 2015. [DOI] [PMC free article] [PubMed]
66.Weisz L, et al. Mutant p53 enhances nuclear factor kappaB activation by tumor necrosis factor alpha in cancer cells. Cancer Res. 2007;67(6):2396–401. [DOI] [PubMed]
67.Cooks T, et al. Mutant p53 prolongs NF-κB activation and promotes chronic inflammation and inflammation-associated colorectal cancer. Cancer Cell. 2013;23(5):634–46. [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1.^{(3.2MB, docx)}

Supplementary Material 2.^{(2.2MB, pdf)}

Data Availability Statement

The sequencing raw data can be accessed from ENA/SRA using accession number PRJEB71525.

[CR1] 1.Borley J, et al. Does tumour biology determine surgical success in the treatment of epithelial ovarian cancer? A systematic literature review. Br J Cancer. 2012;107(7):1069–74. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. Cancer J Clin. 2020;70(1):7–30. [DOI] [PubMed] [Google Scholar]

[CR3] 3.Bast RC Jr., Hennessy B, Mills GB. The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer. 2009;9(6):415–28. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Herzog TJ. The current treatment of recurrent ovarian cancer. Curr Oncol Rep. 2006;8:448–54. [DOI] [PubMed] [Google Scholar]

[CR5] 5.Stuart GC et al. 2010 Gynecologic Cancer InterGroup (GCIG) consensus statement on clinical trials in ovarian cancer: report from the Fourth Ovarian Cancer Consensus Conference. Int J Gynecol Cancer. 2011;21(4):750-5. [DOI] [PubMed] [Google Scholar]

[CR6] 6.Mehrotra M, et al. Drug-Resistant Epithelial Ovarian Cancer: Current and Future Perspectives. Adv Exp Med Biol. 2024;1452:65–96. [DOI] [PubMed] [Google Scholar]

[CR7] 7.Dilawari A, et al. FDA Approval Summary: Mirvetuximab Soravtansine-Gynx for FRα-Positive, Platinum-Resistant Ovarian Cancer. Clin Cancer Res. 2023;29(19):3835–40. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Gupta S, et al. Maintenance therapy for recurrent epithelial ovarian cancer: current therapies and future perspectives – a review. J Ovarian Res. 2019;12(1):103. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Joris S, et al. Efficacy of olaparib in advanced cancers with germline or somatic mutations in BRCA1, BRCA2, CHEK2 and ATM, a Belgian Precision tumor-agnostic phase II study. ESMO Open. 2023;8(6):102041. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR10] 10.Bell D, et al. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474(7353):609–15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Bertelsen BI, et al. Molecular analysis of the PI3K-AKT pathway in uterine cervical neoplasia: Frequent PIK3CA amplification and AKT phosphorylation. Int J Cancer. 2006;118(8):1877–83. [DOI] [PubMed] [Google Scholar]

[CR12] 12.Hahne JC, et al. Downregulation of AKT reverses platinum resistance of human ovarian cancers in vitro. Oncol Rep. 2012;28(6):2023–8. [DOI] [PubMed] [Google Scholar]

[CR13] 13.Ohta T, et al. Inhibition of Phosphatidylinositol 3-Kinase Increases Efficacy of Cisplatin in in Vivo Ovarian Cancer Models. Endocrinology. 2006;147(4):1761–9. [DOI] [PubMed] [Google Scholar]

[CR14] 14.Furman RR, et al. Idelalisib and Rituximab in Relapsed Chronic Lymphocytic Leukemia. N Engl J Med. 2014;370(11):997–1007. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.André F, et al. Alpelisib for PIK3CA-Mutated, Hormone Receptor–Positive Advanced Breast Cancer. N Engl J Med. 2019;380(20):1929–40. [DOI] [PubMed] [Google Scholar]

[CR16] 16.Fritsch C, et al. Characterization of the novel and specific PI3Kα inhibitor NVP-BYL719 and development of the patient stratification strategy for clinical trials. Mol Cancer Ther. 2014;13(5):1117–29. [DOI] [PubMed] [Google Scholar]

[CR17] 17.Konstantinopoulos PA, et al. Primary Analysis of EPIK-O/ENGOT-ov61: Alpelisib Plus Olaparib Versus Chemotherapy in Platinum-Resistant or Platinum-Refractory High-Grade Serous Ovarian Cancer Without BRCA Mutation. J Clin Oncol. 2025;43(26):2908–17. [DOI] [PubMed] [Google Scholar]

[CR18] 18.Shenoy PS, et al. Developing Clinically Relevant Acquired Chemoresistance Models in Epithelial Ovarian Cancer Cell Lines. Bio Protoc. 2022;12(3):e4310. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Gaikwad SM, et al. Non-invasive imaging of phosphoinositide-3-kinase-catalytic-subunit-alpha (PIK3CA) promoter modulation in small animal models. PLoS ONE. 2013;8(2):e55971. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Phadte P, et al. Autophagy-mediated ID1 turnover dictates chemo-resistant fate in ovarian cancer stem cells. J Exp Clin Cancer Res. 2024;43(1):222. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Andrews S. FastQC: a quality control tool for high throughput sequence data. 2010. Available online at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc.

[CR22] 22.Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Li H, Durbin R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics. 2009;25(14):1754–60. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.McKenna A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Van der Auwera GA, et al. From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline. Curr Protoc Bioinf. 2013;43(1110):p11101–111033. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Liu X, et al. dbNSFP v4: a comprehensive database of transcript-specific functional predictions and annotations for human nonsynonymous and splice-site SNVs. Genome Med. 2020;12(1):103. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Zuguang Gu, Roland Eils, Matthias Schlesner. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32(18):2847–9. [DOI] [PubMed]

[CR28] 28.Li H, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25(16):2078–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Koboldt DC, et al. VarScan: variant detection in massively parallel sequencing of individual and pooled samples. Bioinformatics. 2009;25(17):2283–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010;38(16):e164. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Gaikwad SM, et al. Differential activation of NF-κB signaling is associated with platinum and taxane resistance in MyD88 deficient epithelial ovarian cancer cells. Int J Biochem Cell Biol. 2015;61:90–102. [DOI] [PubMed] [Google Scholar]

[CR32] 32.Messeguer X, et al. PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics. 2002;18(2):333–4. [DOI] [PubMed] [Google Scholar]

[CR33] 33.Dhadve AC et al. Decoding molecular interplay between RUNX1 and FOXO3a underlying the pulsatile IGF1R expression during acquirement of chemoresistance. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease. 2020;1866(6):165754. [DOI] [PubMed]

[CR34] 34.Davila JI, et al. Frequent POLE-driven hypermutation in ovarian endometrioid cancer revealed by mutational signatures in RNA sequencing. BMC Med Genomics. 2021;14(1):165. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Briggs S, Tomlinson I. Germline and somatic polymerase ε and δ mutations define a new class of hypermutated colorectal and endometrial cancers. J Pathol. 2013;230(2):148–53. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Hino H, et al. Clinicopathological and mutational analyses of colorectal cancer with mutations in the POLE gene. Cancer Med. 2019;8(10):4587–97. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Voropaeva EN, et al. Clinical aspects of TP53 gene inactivation in diffuse large B-cell lymphoma. BMC Med Genomics. 2019;12(Suppl 2):35. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Kim HS, et al. Direct measurement of engineered cancer mutations and their transcriptional phenotypes in single cells. Nat Biotechnol. 2024;42(8):1254–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Holstege H, et al. High Incidence of Protein-Truncating TP53 Mutations in BRCA1-Related Breast Cancer. Cancer Res. 2009;69(8):3625–33. [DOI] [PubMed] [Google Scholar]

[CR40] 40.Leroy B, Anderson M, Soussi T. TP53 Mutations in Human Cancer: Database Reassessment and Prospects for the Next Decade. Hum Mutat. 2014;35(6):672–88. [DOI] [PubMed] [Google Scholar]

[CR41] 41.Astanehe A, et al. Mechanisms underlying p53 regulation of PIK3CA transcription in ovarian surface epithelium and in ovarian cancer. J Cell Sci. 2008;121(Pt 5):664–74. [DOI] [PubMed] [Google Scholar]

[CR42] 42.Singh B, et al. 53 regulates cell survival by inhibiting PIK3CA in squamous cell carcinomas. Genes Dev. 2002;16(8):984–93. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Tew WP, et al. PARP Inhibitors in the Management of Ovarian Cancer: ASCO Guideline. J Clin Oncol. 2020;38(30):3468–93. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Kadri MSN, et al. Mutational Landscape for Indian Hereditary Breast and Ovarian Cancer Cohort Suggests Need for Identifying Population Specific Genes and Biomarkers for Screening. Front Oncol. 2020;10:568786. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Mannan AU, et al. Detection of high frequency of mutations in a breast and/or ovarian cancer cohort: implications of embracing a multi-gene panel in molecular diagnosis in India. J Hum Genet. 2016;61(6):515–22. [DOI] [PubMed] [Google Scholar]

[CR46] 46.Furet P, et al. Discovery of NVP-BYL719 a potent and selective phosphatidylinositol-3 kinase alpha inhibitor selected for clinical evaluation. Bioorg Med Chem Lett. 2013;23(13):3741–8. [DOI] [PubMed] [Google Scholar]

[CR47] 47.Kolasa IK, et al. PIK3CA amplification associates with resistance to chemotherapy in ovarian cancer patients. Cancer Biol Ther. 2009;8(1):21–6. [DOI] [PubMed] [Google Scholar]

[CR48] 48.Lee S, et al. Activation of PI3K/Akt pathway by PTEN reduction and PIK3CA mRNA amplification contributes to cisplatin resistance in an ovarian cancer cell line. Gynecol Oncol. 2005;97(1):26–34. [DOI] [PubMed] [Google Scholar]

[CR49] 49.McIntyre JB, et al. PIK3CA mutational status and overall survival in patients with cervical cancer treated with radical chemoradiotherapy. Gynecol Oncol. 2013;128(3):409–14. [DOI] [PubMed] [Google Scholar]

[CR50] 50.Arjumand W, et al. Phosphatidyl inositol-3 kinase (PIK3CA) E545K mutation confers cisplatin resistance and a migratory phenotype in cervical cancer cells. Oncotarget. 2016;7(50):82424–39. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR51] 51.Kim KJ, et al. PI3K-targeting strategy using alpelisib to enhance the antitumor effect of paclitaxel in human gastric cancer. Sci Rep. 2020;10(1):12308. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Thakur B, Ray P. Cisplatin triggers cancer stem cell enrichment in platinum-resistant cells through NF-κB-TNFα-PIK3CA loop. J Experimental Clin Cancer Res. 2017;36(1):164. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Jacobs KM, et al. GSK-3β: A Bifunctional Role in Cell Death Pathways. Int J Cell Biology. 2012;2012(1):930710. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Beurel E, Jope RS. The paradoxical pro- and anti-apoptotic actions of GSK3 in the intrinsic and extrinsic apoptosis signaling pathways. Prog Neurobiol. 2006;79(4):173–89. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Gunaratna, RT, et al. Dynamic role of the codon 72 p53 single-nucleotide polymorphism in mammary tumorigenesis in a humanized mouse model. Oncogene. 2019;38(18):3535–50. 10.1038/s41388-018-0630-4. [DOI] [PMC free article] [PubMed]

[CR56] 56.Toulany M, et al. Akt promotes post-irradiation survival of human tumor cells through initiation, progression, and termination of DNA-PKcs-dependent DNA double-strand break repair. Mol Cancer Res. 2012;10(7):945–57. 10.1158/1541-7786.MCR-11-0592. [DOI] [PubMed]

[CR57] 57.Mueck K, et al. Akt1 stimulates homologous recombination repair of DNA double-strand breaks in a Rad51-dependent manner. Int J Mol Sci. 2017;18(11). [DOI] [PMC free article] [PubMed]

[CR58] 58.Tang J, et al. SOX2 recruits KLF4 to regulate nasopharyngeal carcinoma proliferation via PI3K/AKT signaling. Oncogenesis. 2018;7(8):61. [DOI] [PMC free article] [PubMed]

[CR59] 59.Mehta GA, et al. SOX4 and SMARCA4 cooperatively regulate PI3k signaling through transcriptional activation of TGFBR2. npj Breast Cancer. 2021;7(1):40. [DOI] [PMC free article] [PubMed]

[CR60] 60.Yu B, et al. DDX55 promotes hepatocellular carcinoma progression by interacting with BRD4 and participating in exosome-mediated cell-cell communication. Cancer Sci. 2022;113(9):3002–17. [DOI] [PMC free article] [PubMed]

[CR61] 61.Yang N, et al. Transcriptional regulation of PIK3CA oncogene by NF-κB in ovarian cancer microenvironment. PLOS ONE. 2008;3(3):e1758. [DOI] [PMC free article] [PubMed]

[CR62] 62.Astanehe A, et al. The transcriptional induction of PIK3CA in tumor cells is dependent on the oncoprotein Y-box binding protein-1. Oncogene. 2009;28(25):2406–18. [DOI] [PubMed]

[CR63] 63.Wang, Z., et al., PIK3CA Is Regulated by CUX1, Promotes Cell Growth and Metastasis in Bladder Cancer via Activating Epithelial-Mesenchymal Transition. Front Oncol. 2020;10. [DOI] [PMC free article] [PubMed]

[CR64] 64.Kennedy MC, Lowe SW. Mutant p53: it’s not all one and the same. Cell Death Differ. 2022;29(5):983–7. [DOI] [PMC free article] [PubMed]

[CR65] 65.Walerych D, Lisek K, Del Sal G. Mutant p53: One, No One, and One Hundred Thousand. Front Oncol. 2015. [DOI] [PMC free article] [PubMed]

[CR66] 66.Weisz L, et al. Mutant p53 enhances nuclear factor kappaB activation by tumor necrosis factor alpha in cancer cells. Cancer Res. 2007;67(6):2396–401. [DOI] [PubMed]

[CR67] 67.Cooks T, et al. Mutant p53 prolongs NF-κB activation and promotes chronic inflammation and inflammation-associated colorectal cancer. Cancer Cell. 2013;23(5):634–46. [DOI] [PMC free article] [PubMed]

PERMALINK

Alpelisib enhances cisplatin mediated cytotoxicity in non-mutated, PIK3CA-dependent high-grade serous ovarian cancer

Megha Mehrotra

Sourav Chakraborty

Elveera Saldanha

Diksha Joshi

Ajit Dhadve

Asmita Sakpal

Pratik Chandrani

Amita Maheshwari

Santosh Menon

Jaya Ghosh

Sudeep Gupta

Pritha Ray

Abstract

Background

Methods

Results

Conclusion

Supplementary Information

Background

Materials and methods

Cell lines and culture conditions

Site-directed mutagenesis and construction of mutant p53 clonal library

Lentiviral mediated silencing of NFKB and overexpression

Establishment of primary culture of patient-derived cells

Genomic DNA Isolation

Panel-based NGS for identification of germline and somatic mutations

Somatic variant calls and filtration

Germline variant calls and filtration

Copy number assessment

Quantitative real-time PCR

Gene expression profiling using PCR array

Immunoblotting

Nuclear and cytoplasmic fractionation

Immunofluorescence

PIK3CA promoter analysis

Chromatin immunoprecipitation

Cell viability assay

Long-term survival assay

Detection of apoptosis by annexin PI

Boyden chamber assay

Wound-healing assay

Tumour xenograft studies and bioluminescence imaging

Establishment of patient-derived xenograft

TCGA data analysis

Statistical analysis

Results

PIK3CA amplified primary tumor cells exhibit sensitivity to cisplatin and alpelisib combination

Fig. 1.

Table 1.

Mutational spectrum of primary tumour cells

Fig. 2.

Transcriptional and functional characterization of the novel mutant R196Q

Fig. 3.

p53 mutations exhibit a complex regulation of the PIK3CA gene that dictates sensitivity to cisplatin- alpelisib treatment

Fig. 4.

Differential binding of mutant p53 and NFκB modulates PIK3CA activation after cisplatin treatment

Fig. 5.

Cisplatin-alpelisib treatment effectively reduced tumor growth than cisplatin in TOV112D-R175H tumor xenografts

Fig. 6.

Cisplatin-Alpelisib treatment decreases ascites burden in patient-derived xenograft model

Fig. 7.

Discussion

Fig. 8.

Supplementary Information

Acknowledgements

Abbreviations

Authors’ contributions

Data availability

Declarations

Ethics approval and consent to participate

Competing interests

Footnotes

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK