Abstract
Trastuzumab deruxtecan (T-DXd), an antibody–drug conjugate composed of an anti-HER2 antibody and a cytotoxic topoisomerase I inhibitor, is approved for the treatment of HER2-positive (HER2+) and HER2-low breast cancer as well as HER2-high gastric and HER2-mutant lung cancer tumors. The AKT inhibitor capivasertib is approved for the treatment of HER2− estrogen receptor–positive breast cancer with alterations in PIK3CA, PTEN, and AKT-1. The potential for the combination of T-DXd with AKT inhibition to enhance antitumor activity was explored in HER2+ or HER2-low preclinical models. In vitro, combination activity was observed in both HER2-high– and HER2-low–expressing breast cancer as well as in gastric, endometrial, and ovarian models, irrespective of HER2 expression level or PI3K–AKT status pathway alterations. The T-DXd–capivasertib combination effect translated in vivo with increased antitumor benefit in HER2-expressing, PI3K–AKT pathway–altered tumor xenografts when compared with the combination of trastuzumab and capivasertib. In cell lines sensitive to the combination, combining T-DXd with capivasertib targeted complimentary pathways which resulted in disruption of the cell cycle and increased cell death. These results suggest that T-DXd combined with capivasertib has the potential to be active in HER2+ as well as HER2-low tumors independent of PI3K pathway alteration status.
Introduction
Members of the PI3K–AKT pathway are among the most frequently mutated genes in cancer, with alterations in PIK3CA, PTEN, and AKT1 found in many diseases, including breast, prostate, gastric, ovarian, endometrial, lung, and head and neck cancers. PI3K–AKT signaling plays a pivotal role in many key cellular processes such as cell growth, survival, and gene expression that are critical to tumor progression, with activation of the pathway playing an important role in many tumors, often in co-operation with other tumor drivers (1). AKT signaling is also important in cell stress responses (2). As a result, combinations with AKT or PI3K signaling inhibitors have been shown to enhance efficacy of many other agents, including chemotherapy, radiotherapy, hormonal therapy, and signaling inhibitors (3–5).
The AKT inhibitor capivasertib (6) was recently approved for use in combination with fulvestrant for treatment of PIK3CA/AKT1/PTEN-altered advanced HR-positive breast cancer based on the results of the CAPItello291 trial (7). In preclinical models of estrogen receptor–positive (ER+) breast cancer, capivasertib combines with fulvestrant to enhance antitumor activity in both CDK4/6 treatment-naïve and CDK4/6 inhibitor–treated tumor models (4, 8) and also combines with the next-generation selective ER degraders camizestrant and palbociclib to enhance the antitumor effects across a broad range of ER+ tumor models (8). More broadly capivasertib has been shown to combine with androgen receptor inhibitors, taxane-based chemotherapy, and radiation in preclinical models (5, 9–13). Therefore, targeting AKT has potential to enhance the effectiveness of combination partners.
Trastuzumab deruxtecan (T-DXd/DS-8201a/Enhertu) is an antibody–drug conjugate (ADC) composed of an anti-HER2 antibody, a cleavable tetrapeptide-based linker, and a cytotoxic topoisomerase I inhibitor (1). T-DXd has demonstrated clinical activity in patients with breast cancer with a wide range of HER2 expression levels. T-DXd is approved for the treatment of metastatic HER2-positive (HER2+; IHC 3+ and 2+/ISH+), HER2-low (IHC1+ and 2+/ISH−), and HER2 ultralow (IHC 0) breast cancers (14–16). Based on these data, it has become a new standard of care for this stage of the disease. T-DXd has also shown benefit in HER2+ tumors across other disease settings, including gastric, ovarian, endometrial, and lung cancers, leading to a tumor-agnostic approval (17–19).
When activated through homodimerization or heterodimerization, HER2 can signal through the PI3K–AKT pathway. Breast tumors with amplified HER2 are sensitive to HER2 signaling inhibition with trastuzumab, and preclinical models show HER2+ breast tumors are responsive to PI3K and AKT inhibitors (20, 21). Targeting AKT with capivasertib reduces growth of HER2+ breast and gastric tumor cells and combines with trastuzumab or AZD8931 (a pan-ERBB inhibitor) to increase antitumor effects. Preclinically, combined treatment with capivasertib and AZD8931(EGFR/HER2/HER3 inhibitor) results in synergistic growth inhibition and enhanced cell death in breast cancer models, specifically in HER2-amplified models (6, 22). Although modest inhibition of the AKT pathway by T-DXd in vitro has been previously demonstrated in HER2+ breast cancer cell lines (23), the combination of T-DXd and capivasertib and cellular consequences have not been explored in preclinical models. This is of particular interest for a number of reasons. First T-DXd and capivasertib are commonly used to treat different but overlapping segments, T-DXd (ER± HER2 low to high expression) and capivasertib ER+ PIK3CA/PTEN/AKT-altered breast cancer. Second, it is not clear whether T-DXd combined with capivasertib is more effective that the combination of trastuzumab and capivasertib in HER2-high or -dependent tumors. Third, PIK3CA alterations are associated with poor prognosis and resistance to HER2-targeted agents, including trastuzumab and pertuzumab (24, 25). In the phase III DESTINY-Breast09 study, PIK3CA alterations reduced the effectiveness of the taxane plus trastuzumab and pertuzumab combination; however, the combination of T-DXd and pertuzumab was less affected by PIK3CA mutations (24, 26). Therefore, we sought to determine whether combining capivasertib with T-DXd would provide improved benefit compared with targeting HER2 alone. To address these questions, the combinations were assessed by screening both breast cancer–focused and broad cell panels, as well as a range of tumor xenograft models which demonstrated enhanced activity of the combination of capivasertib and T-DXd.
Alterations in the PI3K/AKT pathway are frequent across multiple cancers, including ovarian cancer, endometrial cancer, and all subtypes of breast cancer. Aberrant PI3K/AKT activation is a key driver of multidrug resistance via several mechanisms. These include modulation of drug transport across the cell membrane by upregulating efflux pumps, particularly ATP-binding cassette (ABC) transporters (27). Several chemotherapeutics—such as topoisomerase I inhibitors, including irinotecan and SN-38,—are substrates of these transporters (28). In addition, abnormal PI3K/AKT signaling can negatively affect chemotherapy-induced apoptosis by enhancing survival pathways and altering apoptotic gene expression—upregulating antiapoptotic genes (e.g., BCL-2 and XIAP) while downregulating proapoptotic genes (e.g., BAX; ref. 27). Furthermore, PI3K pathway aberrations have been linked to resistance to HER2-targeted agents, including trastuzumab (21, 29).
Given the relevance of PI3K pathway activation to both the T-DXd payload and antibody mechanism, and its high prevalence across breast cancer subtypes, combining T-DXd with PI3K–AKT pathway inhibitors may enhance T-DXd efficacy and delay or reduce the onset of resistance.
Materials and Methods
Cell lines
Cell lines used in this study, along with their sources and growth conditions, are shown in Supplementary Table 1. All cell lines were authenticated by short-tandem repeat analysis and screened by PCR for Mycoplasma. Cells were used in studies between passages 2 to 10.
In vitro cell viability combination screen
For combination screening, cell viability was assessed using a CellTiter-Glo (CTG) endpoint. Cells were seeded at a density of 200 to 2000 cells/well, depending on doubling time, into 384-well white solid bottom plates (Greiner 781080) in 60 μL RPMI-1640 media containing 2 mmol/L L-glutamine and 10% (v/v) fetal bovine serum (FBS). Assay plates were incubated overnight (37°C; 5% CO2) before treating with compounds or dimethyl sulfoxide (DMSO) the next day (day 0). The treatment was performed by dispensing onto cells in a 6 × 6 concentration–response matrix format using the Echo 555 acoustic liquid handling system. A control viability count was also established on day 0 by reading a plate of untreated cells. Compound-treated plates were incubated for 7 days (37°C; 5% CO2), and cell viability was measured using the CTG viability assay (G7573, Promega). Briefly, 30 μL of CTG reagent was added to each well, incubated for 30 minutes at ambient temperature, and luminescence was measured using a Tecan F500 microplate reader (Promega). Data were normalized to the original day 0 untreated controls and the maximum day 7 growth (DMSO-only treated wells). Combination analysis resulting in the “matrix plot” heat maps was performed using GeneData Screener (RRID: SCR_022506).
Cell lysate generation and protein quantification for Western blot analysis
1 × 106 cells were seeded into 10-cm tissue culture–treated dishes in 10 mL RPMI-1640 media containing 2 mmol/L L-glutamine and 10% FBS. Cells were incubated (37°C; 5% CO2) for 24 hours. After this period, media were aspirated and replaced with 10 mL of fresh media containing the treatments. Compound-treated plates were then incubated (37°C; 5% CO2) for 24, 48, and 72 hours.
At the appropriate time points, cells were washed twice with cold phosphate-buffered saline (PBS). Cells were then lysed on ice for 30 minutes by adding 100 µL of RIPA buffer (89901, Themo Fisher Scientific) supplemented with 1 µL of Halt protease and phosphatase inhibitor cocktail (186128, Thermo Fisher Scientific) and 5 µmol/L ethylene diamine tetra-acetic acid (1861274, Thermo Fisher Scientific). Cell lysates were centrifuged at 4°C, 13,000 rpm for 10 minutes. Supernatants were collected and quantified for protein using the Pierce BCA Protein Assay Kit (23225, Thermo Fisher Scientific) according to the manufacturer’s instructions.
Western blot analysis
For Western blot analysis, samples were normalized to equivalent protein concentrations by diluting in RIPA buffer and adding Sample Reducing Agent (NuPAGENP0009, Thermo Fisher Scientific) and 1 × NuPAGE LDS Sample Buffer (NuPAGENP007, Thermo Fisher Scientific). Samples were then boiled at 95°C for 5 minutes, followed by brief centrifugation at 13,000 rpm before loading 20 μg onto NuPAGE 4% to 12% Bis-Tris Midi Gels (WG1402B, Thermo Fisher Scientific). Gels were run in NuPAGE MOPS SDS Running Buffer (NP0001, Thermo Fisher Scientific) at 120 V for 2 hours. Transfer was performed using an iBlot (Thermo Fisher Scienitific) semi-dry transfer system at 20 V for 11 minutes. Blots were blocked for 1 hour at ambient temperature in a blocking buffer consisting of 5% (w/v) nonfat dry milk in 0.05% (v/v) Triton X-100 in Tris-buffered saline (TBS-T). Blots were probed with primary antibodies at 4°C overnight, followed by incubation with horseradish peroxidase–conjugated secondary antibody at ambient temperature for 2 hours. Blots were analyzed using Pierce West Dura and Femto Dura chemiluminescent reagents on the G:Box imager (manual exposure setting) with SynGene software. A list of antibodies and dilutions used in this study is shown in Supplementary Table S2.
To detect total protein, membranes previously probed for the phosphoprotein were stripped using Restore Western Blot Stripping Buffer (Thermo Fisher Scientific, 21059). After the initial phosphoprotein detection, blots were washed three times in TBS (5 minutes each) and then incubated in stripping buffer for 15 to 20 minutes at room temperature with gentle agitation. Membranes were subsequently washed in TBS three times (10 minutes each), reblocked in 5% BSA in TBS-T, and incubated overnight at 4°C with primary antibody against the total protein.
Cell-cycle analysis
1 × 106 KPL4 cells were seeded into 10-cm tissue culture–treated dishes in 10 mL RPMI media containing 10% (v/v) FBS + 2 mmol/L L-glutamine and were incubated (37°C; 5% CO2) for 24 hours. After this period, media were aspirated and replaced with 10 mL fresh RPMI media containing 10% FBS +2 mmol/L L-glutamine and capivasertib and/or T-DXd.
After 24, 48, and 72 hours of incubation (37°C; 5% CO2), cells were harvested, washed with PBS, and fixed by adding 1 mL cold 70% (v/v) paraformaldehyde dropwise while vortexing. Cells were then incubated on ice for 30 minutes. Following this, cells were centrifuged at 13,000 rpm for 5 minutes, and the supernatant was aspirated. The cell pellets were washed with PBS and treated with 50 µL RNase A (EN0531, Thermo Fisher Scientific). Subsequently, 400 L of propidium iodide solution (1 mg/mL, P3566, Thermo Fisher Scientific) was added per million cells, and the mixture was incubated for 5 to 10 minutes at ambient temperature. Samples were analyzed on the BD FACS Calibur (BD Biosciences), collecting a total of 50,000 to 100,000 events. FL2-A was plotted as a histogram and gates established to calculate the proportion of cells in each stage of the cell cycle (G1, S, and G2–M). Data were processed in FlowJo v10.8 (RRID: SCR_008520), applying the Dean–Jett–Fox model.
Horizon combination screen
Capivasertib, T-DXd, and exatecan were screened as monotherapy or in combination in Horizon Discovery’s OncoSignature Cell Panel, which includes 300 cell lines covering 18 indications. The screen was performed according to the standard protocol (https://www.revvity.com/gb-en/category/cell-panel-screening). Cells were seeded into 384-well tissue culture–treated plates in 25 µL of growth media. Cells were equilibrated in assay plates via centrifugation and incubated (37°C; 5% CO2) for 24 hours before compound treatment. Combinations were dosed in a 10 × 2 matrix format (9 doses of T-DXd or exatecan ± 1 dose of capivasertib). At the time of compound treatment, a set of untreated assay plates were collected to measure ATP levels. This was done by adding CTG 2.0 reagent (Promega) and reading the luminescence on Envision plate readers (Perkin Elmer). Compounds were transferred to assay plates using Echo Acoustic liquid Handling Systems (Labcyte). Assay plates were incubated with compound for 6 days and then analyzed by the addition of CTG 2.0 reagent. All data points were collected via automated processes and were subject to quality control and analyzed using Horizon’s proprietary software. All analyses for combination screening data following initial processing with Horizon’s software were performed using R version 4.3.1 “Beagle Scouts”.
The correlation of highest single agent (HSA; exatecan and capivasertib, T-DXd and capivasertib; refs. 30, 31) with HER2 gene expression: For all solid tumor cell lines included in the Horizon panel, a Spearman correlation was calculated between HSA (exatecan in combination with capivasertib, T-DXd in combination with capivasertib) and ERBB2 log2-transformed gene expression (transcripts per million) levels. Correlation of Emax with HSA: For all solid tumor cell lines included in the Horizon panel, a Spearman correlation was calculated between HSA (exatecan in combination with capivasertib, T-DXd in combination with capivasertib) and combination Emax. To assess the association with PI3K–AKT pathway mutation status across all solid tumor cell lines, an association between PI3K pathway mutations and HSA (exatecan and capivasertib, T-DXd and capivasertib) was assessed using nonparametric Wilcoxon rank-sum test. A P value of <0.05 was considered significant. All solid tumor cell lines with at least one nonsilent mutation in PIK3CA, AKT-1, AKT-2, and AKT-3 genes or a deletion in PTEN gene were classified as mutant.
A similar association was carried out for breast cancer lines: For all breast tumor cell lines included in the Horizon panel, an association between PI3K pathway mutations and HSA (exatecan and capivasertib, T-DXd and capivasertib) was assessed using nonparametric Wilcoxon rank-sum test. A P value of <0.05 was considered significant. All breast tumor cell lines with at least one nonsilent mutation in PIK3CA, AKT-1, AKT-2, and AKT-3 genes or a deletion in PTEN gene were classified as mutant. The size of the dots in the box plot was proportional to the increasing ERBB2 copy number in the breast cancer lines, used to examine whether mutant cell lines were enriched for higher ERBB2 copy number.
In vivo tumor models
Female BALBc nude mice were implanted subcutaneously with KPL-4, JIMT-1, AN3CA, and SKOV3 cell lines. Tumors were measured using calipers, and tumor volume (TV) was calculated as TV= (length × width2)/2. Tumor growth inhibition (TGI) was calculated as %TGI = (1-TVT/TVC) × 100%, in which TVT is the mean TV of the treatment group and TVC is the mean TV of the control group. T-DXd was administered intravenously every 3 weeks at either 3 mg/kg or 10 mg/kg. Capivasertib was administered by oral gavage at 130 mg/kg using a 4 days on, 3 days off cycle. Trastuzumab was administered twice weekly intraperitoneally at either 4 mpk (KPL-4 and JIMT1) or 30 mpk (AN3CA and SKOV3). Irinotecan was administered at 25 mg/kg intraperitoneally once weekly. All formulated materials remained sterile and protected from light. KPL-4 and JIMT-1 models were executed courtesy of Axis Biosciences. AN3CA and SKOV3 models were executed courtesy of Medicilon. All mice were given free access to food and water throughout the course of the study.
All experiments followed the principles of good statistical practice, as well as the PREPARE and ARRIVE guidelines. All animal studies were conducted by contract research organizations in accordance with local authorities, guidelines of the Animal Welfare Act, and the AstraZeneca Global Bioethics policy. Animal studies were performed in accordance with protocols approved by the Medicilon “Institutional Animal Care and Use Committee” along with the AstraZeneca’s “Platform for Animal Research Tracking and External Relationships” (PARTNER) group.
Results
Capivasertib increases sensitivity to T-DXd in HER2+ and HER2-low breast cancer cell lines
To explore the potential for capivasertib to combine with T-DXd, a focused cell panel drug combination screen was performed in a panel of HER2+ and HER2-low cell lines containing 27 breast cancer cell lines (both ER+ and ER−; Fig. 1). A HER2+ gastric cancer cell line NCI N87 was included as a comparator (Supplementary Table 3). Cells were incubated with a range of concentrations of T-DXd (64 nmol/L/10 µg/mL top concentration), trastuzumab (10 µg/mL top concentration), or the active payload analog exatecan (10 nmol/L top concentration) alone or in combination with capivasertib (3 µmol/L top concentration) in a matrix format cell viability combination assay. Cell lines exhibited a range of sensitivities to each monotherapy treatment (Supplementary Table S4). At the highest concentration (3 µmol/L), capivasertib single-agent treatment led to cell death (Emax >100) in three ER+ breast cancer cell lines (EFM19, BT474, and CAMA1) and one triple-negative breast cancer (TNBC) cell line (HCC70). All 3 ER+ breast cancer cell lines had alterations in the PI3K–AKT pathway. Exatecan induced cytotoxicity at 10 nmol/L in 16 cell lines. Following trastuzumab treatment, only BT474 (ER+ HER2+) cells had an Emax above 100 (cytotoxicity) at 10 µg/mL, although BT474 cells were insensitive to T-DXd and exatecan due to a mutation in the TOP1 gene which encodes topoisomerase 1 the target of exatecan.
Figure 1.
In vitro benefit of capivasertib plus T-DXd combination in cell lines. A, Plot of the HSA score vs. combination Emax in 27 breast cancer cell lines and one gastric cancer cell line screened with the capivasertib plus T-DXd over 7 days. The y-axis indicates the combination Emax, reflecting the highest value obtained in the combination matrix. Values between 0 and 100 show growth inhibition, whereas values above 100 indicate cell death. The x-axis displays HSA scores. Different shaped symbols signify the presence of PIK3CA and/or PTEN alterations as indicated. Blue shapes represent HER2-high–expressing cell lines, and red shapes represent HER2-low–expressing cell lines. B, (i) Dose–response and (ii) HSA excess combination matrices for in vitro combination activity in KPL4 (HER2-high) and EFM19 (ER+ HER2-low) breast cancer cell lines and NCI-N87 (HER2-high) gastric cell line. The x-axis represents T-DXd, and the y-axis the capivasertib dose responses. Values in the dose–response matrix show normalized cell viability signals on day 7, compared with DMSO control on day 0. Values between 0 and 100 indicate % growth inhibition, and values above 100 indicate cell death. The HSA excess matrix values represent excess calculated using the HSA model. C, T-DXd and capivasertib single-agent activity in 27 breast cancer cell lines and one gastric cancer cell line vs. the combination of capivasertib plus T-DXd. The y-axis represents monotherapy capivasertib Emax (effect on cell viability at the highest dose), and the x-axis represents the T-DXd Emax. Symbol shapes signify the presence of PIK3CA and/or PTEN alterations. Blue-colored symbols represent cell lines showing combination benefit over single agents (combination Emax >100 and HSA >5), red-colored symbols represent cell lines showing strong monotherapy activity (combination Emax >100 and HSA <5), and yellow represents cell lines showing minimal monotherapy and no combination benefit (combination Emax <100 and HSA <5). D, (i) Dose–response and (ii) HSA excess combination matrices for human primary bone marrow–derived CD34+ cells differentiated into erythroid, myeloid, or megakaryocyte lineages. Due to cell differentiation over time, a day 0 control was not calculated; therefore, values are normalized to DMSO control only, with values between 0 and 100 indicating combined % growth inhibition and cell death. The x-axis represents T-DXd, and the y-axis represents the capivasertib dose responses.
The combination treatment of capivasertib plus T-DXd resulted in strong combination benefit (combination Emax >100 and HSA score above 5) in seven of 27 breast cancer cell lines and the gastric cell line NCIN87 across this small panel (Fig. 1A; Supplementary Fig. S1; Supplementary Table S3). Enhanced combination activity was observed in PI3K pathway–altered (PIK3CA and PTEN) and nonaltered cells. Five of eight responders had PIK3CA or PTEN alterations, and 3 were PIK3CA and PTEN wild-type (Supplementary Table S1).
In the remaining 20 breast cancer cell lines, an increase in cell death or growth inhibition (combination Emax) compared with single-agent treatment and HSA scores between 0.1 and 5 were observed in 18 cell lines, indicating more modest combination activity. Only BT474 had no combination activity despite this cell line being sensitive to HER-2 inhibition. This is consistent with a lack of topoisomerase 1 sensitivity as BT474 showed no combination benefit from exatecan alone (Supplementary Fig. S1). One other cell line (HCC1954) had an HSA score below 0, indicating an antagonistic interaction between the two agents. Although combination benefit was greater in six HER2+ cell lines (HCC1419, KPL4, ZR-75-30, HCC1569, AU565, and SKBR3), activity was also observed in the HER2-low cell line EFM-19 (determined by HER2 mRNA expression level; Fig. 1A and B). Overall, the combination of T-DXd and capivasertib produced more than a 20% increase in growth inhibition and/or cell death in 11 cell lines at a clinically relevant capivasertib concentration of 1 µmol/L, irrespective of PI3K pathway status and HER2 level (Supplementary Fig. S2). However strong combination benefit scores were only observed in cell lines with modest sensitivity to both agents and not in any model insensitive to either agent (Fig. 1C).
In the same cell panel, exatecan monotherapy was highly effective above approximately 1 nmol/L, leading to significant cell death. Therefore, the enhancement of maximal cell kill (Emax) in combination with capivasertib was less effective, although modest combination activity was observed in 4 of 28 cell lines (KPL4, EFM19, NCIN87, and HCC1569) at lower concentrations of exatecan and strong combination benefit was observed in HER2-low ER+ breast cancer cell line MDA-MB-361 (HSA 9.1) (Supplementary Figs. S3 and 4). Consistent with a lack of topoisomerase 1 activity, BT474 showed no combination benefit from exatecan (Supplementary Figs. S3 and 4). Collectively these data demonstrate that capivasertib can combine with T-DXd to increase antitumor effects and that this combination benefit is observed in cell lines that have both high and low HER2 expression as well as in cell lines with and without alterations in PIK3CA or PTEN.
Combining T-DXd and capivasertib does not increase in vitro bone marrow toxicity at clinically relevant concentrations
Combination activity between canonical signaling inhibitors and cytotoxic agents can commonly increase cytotoxicity in normal tissues, with bone marrow often showing increased sensitivity to combinations. To explore the potential impact on normal tissues and specifically the potential impact on bone marrow, the combination activity of capivasertib plus T-DXd was examined in human primary bone marrow–derived CD34+ cells differentiated into erythroid, myeloid, or megakaryocyte lineages. All lineages were sensitive to T-DXd at clinically achievable concentrations of 133 μg/mL (0.9 µmol/L; Fig. 1D). This effect is consistent with clinical observations in which multi-lineage effects are observed despite a lack of HER2 expression in the bone marrow (32). At the highest doses of capivasertib, there was a modest monotherapy effect on progenitor cells (Fig. 1D); however, no synergistic combination activity in the 2D in vitro bone marrow assay was observed (Supplementary Fig. S5). Lack of combination effect, even at higher doses of both agents, suggests the combination has a low risk of exacerbating hematologic toxicity; however, it was not possible to assess other normal tissue effects in the context of in vitro and in vivo models.
T-DXd and capivasertib act independently to increase antiproliferative effects and cell death in sensitive cell lines
The mechanistic effects of the capivasertib T-DXd combination on pharmacodynamic biomarkers was examined in HER2-amplified KPL4 and HER2-low EFM19 (ER+ PIK3CA pH1047L) breast cancer cell lines and the HER2-low endometrial cell line AN3CA (PTEN-mutant). Cells were treated with clinically relevant concentrations of T-DXd (Cmax) at 100 µg/ml T-DXd and capivasertib at 1 µmol/L or the combination for 4, 24, 48, and 72 hours. The effect of each monotherapy and combination treatment on expression of PI3K–AKT, apoptosis, and cell-cycle pathway protein markers were measured by Western blot analysis (Fig. 2; Supplementary Fig. S6). In all three cell lines, monotherapy or combination treatment with capivasertib increased pAKT (S 473) and decreased pPRAS40 (Thr 246), pS6 (S240/244), pGSK3B (S9), and p4EBP1 (S65) levels consistent with the mode of action of capivasertib (Fig. 2A; refs. 2, 4, 6). Monotherapy T-DXd treatment also resulted in a very modest decrease in pPRAS40 (Thr 246) and p4EBP1 (S65) in all three cell lines at later time points at 72 hours (Fig. 2A). This is consistent with T-DXd delivering some PI3K–AKT pathway via inhibition of HER2 signaling T-DXd at the doses used (23). Collectively, these data suggest that modulation of PI3K–AKT signaling is consistent between the monotherapy capivasertib and combination groups. The effect on DNA damage response (DDR) pathway biomarkers associated with T-DXd treatment was also assessed. TOP1 inhibitors induce formation of topoisomerase I cleavable complex (TOP1cc), degradation of TOP1, and induction of DNA damage or cell stress (33). T-DXd alone or in combination with capivasertib elevated levels of the DDR markers pChk1 (S345), pChk2 (Thr 68), pATM (S1981), pATR (S1989), and p-γH2AX as early as 24 hours. Increased p-γH2AX (S139) and cleaved PARP levels were observed in all 3 cell lines treated with T-DXd alone or in combination with capivasertib, indicating increased DNA damage and apoptosis, with similar levels in T-DXd monotherapy and combination-treated cells (Fig. 2B). Total p-γH2AX levels increased in response to T-DXd or the combination, whereas total Chk1/2, ATR, and ATM levels remained unchanged, except in the HER2-low cell line EFM19, in which total Chk1 decreased with T-DXd or the combination (Supplementary Figure 6B). Interestingly, longer-term (72 hours) capivasertib monotherapy treatment had a modest effect on p-γH2AX in KPL4 cells but not in the other cell lines. Increases in p-γH2AX (S139) and cleaved PARP were also observed in the HER2+ gastric cells NCI-N87, used as a control in the cell-panel screen, following treatment with T-DXd alone or in combination with capivasertib (Supplementary Figure 7).
Figure 2.
Western blot analysis of HER2-low (AN3CA and EFM19) and HER2+ (KPL4) cell lines treated with capivasertib alone or in combination with T-DXd. Changes in expression of (A) PI3K–AKT pathway, (B) DDR pathway, (C) cell-cycle pathway, and (D) HER/EGFR pathway protein markers in the HER2+ breast cancer cell line, KPL4, the HER2-low ER+ breast cell line, EFM19, and the HER2-low endometrial cancer cell line, AN3CA, treated with vehicle (DMSO), 1 µmol/L capivasertib, or 100 µg/ml T-DXd monotherapy and the combination for 4, 24, 48, and 72 hours.
To assess effects on the cell cycle, a range of biomarkers were examined: pcdc2(Tyr15), pH3(S10), pRB(S780), and cyclin D1. A decrease in cyclin D1 was observed at 4 hours in KPL4 and EFM19 cells treated with capivasertib, consistent with capivasertib inducing G1/S arrest in sensitive cells. A reduction in cyclin D1 was also seen at later time points in T-DXd–treated AN3CA and EFM19 cells, possibly as a result of an indirect effect on cyclin D1 following S/G2 arrest (Fig. 2C).
Increased S/G2 arrest markers, including pCdc2, pChk1/2, pATM, and pATR were, observed in samples treated with T-DXd alone or in combination with capivasertib but were not enhanced by the combination (Fig. 2C). Following T-DXd treatment, HER2 protein levels were downregulated in EFM19 cell, whereas HER3 was upregulated in ANC3A cells. In addition, modest increases in pERK were observed in KPL4 and EFM19 cells treated with T-DXd and capivasertib. However, no consistent changes in HER2, HER3, and ERK signaling were observed in all the cell lines (Fig. 2D). Overall, no consistent changes in PI3K–AKT or DDR pathway markers was observed following combination treatment compared with either single agent. This suggests that capivasertib and T-DXd target complimentary pathways, mainly acting independently to increase antiproliferative effects or cell death in HER2-low and HER2+ cell lines through dual inhibition of the PI3K pathway and induced DNA damage.
T-DXd in combination with capivasertib increases S and G2 cell populations
Given the combination affected the cell cycle but did not consistently increase apoptosis or DNA damage markers compared with monotherapy treatment, the impact on the cell cycle was assessed. The HER2-amplified cell line KPL4 and HER2-low cell lines AN3CA and EFM19 were treated with 1 µmol/L capivasertib and 100 µg/ml T-DXd alone or in combination for 24 and 48 hours, and cell-cycle distribution was analyzed by flow cytometry (Fig. 3; Supplementary Fig. S8). An increase in the G1 population was observed at 24 hours following capivasertib treatment and persisted up to 48 hours in the HER2-low and PI3K pathway–altered cell lines EFM19 and AN3CA. An increase in the S phase population was observed in all three cell lines treated with T-DXd compared with DMSO at 24 and 48 hours (Fig. 3; Supplementary Fig. S8). This increased further in KPL4 samples treated with capivasertib plus T-DXd versus the controls at 24 hours (70.4% compared with 48.2%; Fig. 3) and in AN3CA samples at 48 hours (89.1% compared with 79.3%; Supplementary Fig. S8). However, this increase was not observed in EFM19 at 24 and 48 hours. Arrest was greater in samples treated with T-DXd combined with capivasertib. These findings suggest that accumulation of S/G2 cell-cycle arrest by T-DXd alone or in combination with capivasertib can lead to cell death which may not be through classical apoptosis pathways. Interestingly each of these cell lines have different proliferation rates—KPL4 approximately 30 hours (34), AN3CA approximately 45 to 50 hours, and EFM19 approximately 81 hours (35). Differences in cell doubling time may therefore reveal a stronger effect at the level of cell-cycle state in rapidly dividing cells. However, this also further supports the observation that the combination induces increased DNA damage and cell death independent of the effect of each monotherapy at a specific point in the cell cycle.
Figure 3.
Active cell-cycle distribution analysis of KPL4, AN3CA, and EFM19 cells treated with T-DXd, capivasertib monotherapy, or the combination. Cell-cycle phase distribution analysis of propidium iodide–labeled cells by flow cytometry after treatment with (A) DMSO, (B) 100 µg/ml T-DXd, (C) 1 µmol/L capivasertib, or (D) the combination for 24 hours.
Combination benefit of capivasertib plus T-DXd does not depend on high HER2 expression or PI3K pathway alterations in a pan tumor cell panel
Having established the combination potential using a focused panel of cells lines, the effects of the capivasertib–T-DXd combination was tested a larger panel of cell lines representative of a much more diverse array of tumor types. This bigger panel screen gives greater insight into potential biomarkers or mutations associated with sensitivity to the combination. A panel of 300 cancer cell lines (252 solid and 48 hematologic) were treated with T-DXd or exatecan monotherapy or in combination with 750 nmol/L capivasertib for 6 days. Combination benefit [defined as combination Emax >100 and HSA score >0.1 (36)] was not limited to a specific cancer type, with activity in cell lines representative of various tumor types, including breast, ovarian, endometrial, gastric, and lung cancer cell lines (Fig. 4A and B; Supplementary Fig. S9). Consistent with the initial cell panel screen, HSA combination benefit was observed in HER2-high– and HER2-low–expressing breast cancer cell lines. A positive trend but no significant correlation was observed between HER2 expression and combination benefit measured by HSA (R = 0.059; Fig. 4C). Monotherapy response to capivasertib can be enriched in cell lines with alterations in PIK3CA, PTEN, or AKT1 (6). No significant association was found between PI3K–AKT pathway alterations (PIK3CA mutations E542 K, E545 K, H1047R, H1047L, and PTEN deletion) and combination benefit (HSA; Fig. 4D).
Figure 4.
Association between ERBB2 expression level or PI3K–AKT pathway alterations and benefit from combination treatments in a cross-tumor type cell panel. A, Representation of Emax vs. HSA scores in 252 cell lines across 18 solid tumor indications. Cells were treated with a concentration range of T-DXd (maximum concentration 100 μg/mL) in the absence or presence of 750 nmol/L capivasertib for 5 days. Dots represent individual cell lines, with colors delineating cancer types as indicated in the key. B, HSA scores for the capivasertib plus T-DXd combination treatment in each of the 300 cell lines in the cell panel. Colors represent cancer types as indicated in the key. C, Correlation plot between HER2 expression levels and the antiproliferative effects of the capivasertib plus T-DXd or capivasertib plus exatecan combinations. Plots indicate Spearman correlation calculated between the HSA (exatecan plus capivasertib, T-DXd plus capivasertib combinations) and the log2-transformed HER2 gene expression. Gray-shaded area represents the 95% confidence interval around the fitted linear model. D, Association between the benefit of capivasertib plus T-DXd or exatecan combination and PI3K pathway mutations. The association between these combinations and HSA (exatecan plus capivasertib, T-DXd plus capivasertib) was evaluated using the nonparametric Wilcoxon rank-sum test. A P value of <0.05 was considered significant. All solid tumor cell lines with at least one nonsilent mutation in PIK3CA, AKT-1, AKT-2, and AKT-3 genes or a PTEN gene alteration or deletion were classified as mutant. Box plot represents median HSA score with 25% and 75% quartiles, and whiskers represent the 1.5*IQR. E, Combination Emax vs. HSA scores in 22 breast cancer cell lines from the large cell panel screen, treated with a concentration range of T-DXd (maximum concentration 100 μg/mL) in the absence or presence of 750 nmol/L capivasertib for 5 days. Colors indicate PIK3CA or PTEN alterations, shapes denote breast cancer subtypes, and sizes represent ERBB2 mRNA expression levels. F, Plot of the HSA score for the capivasertib plus exatecan vs. capivasertib plus T-DXd combination for PI3K pathway WT vs. altered (PIK3CA and PTEN) in 22 breast cancer cell lines. ER and HER2 status are indicated in key. Box plot represents median HSA score with 25% and 75% quartiles, and whiskers represent the 1.5*IQR. IQR, interquartile range.
Response in breast, gastric, and endometrial/uterine subgroups of cell lines was also examined. In the breast cancer subgroup, HSA >0 and combination Emax above 100 (cell death) was observed in 9 of 22 cell lines which included HER2+, ER+ HER2-low, and TNBC cell lines with or without PI3K–AKT pathway alterations. The strongest HSA was observed in 4 ER+ HER2-low breast cancer cell lines T47D (PIK3CA H1047R), MCF7 (PIK3CA E545 K), MDAMB175V II, and CAMA1 (PTEN loss; Fig. 4E). In addition, combination benefit was observed in 10 breast cancer cell lines treated with capivasertib plus T-DXd (Fig. 4E). No significant association was observed between PI3K pathway alterations and combination activity (HSA; P = 0.45; Fig. 4F).
Three of 18 gastric and 1 of 13 endometrial/uterine cancer cell lines treated with capivasertib plus T-DXd showed combination Emax >100 (cell death) and HSA >0, respectively. This number was higher in the capivasertib plus exatecan groups (nine for gastric and seven for endometrial/uterine subgroups). However, five of seven endometrial/uterine cancer cell lines showing combination benefit in the capivasertib plus exatecan group also showed HSA above 0 when treated with combination of capivasertib plus T-DXd, although this effect only led to increased growth inhibition (combination Emax <100), not cell death (Supplementary Fig. S10A and S10B). Hence, no significant association was observed between PI3K–AKT pathway alterations and combination activity (HSA) in the gastric and endometrial/uterine cancer cell lines screened in this study (P = 0.5 and 0.7, respectively; Supplementary Fig. S10C and S10D). Collectively these data show the potential of the capivasertib plus T-DXd combination benefit in HER2-low breast cancer tumors, particularly ER+ segment in addition to HER2+ tumors irrespective of PI3K–AKT pathway alterations. Potential combination benefit was also observed in gastric and endometrial cancer cell lines, although this effect was not as profound as that observed in breast cancer cell lines.
Capivasertib plus T-DXd combination enhances the depth and durability of response in xenograft models
To validate the combination in vivo, the combination of T-DXd and capivasertib was evaluated in four tumor xenograft models, KPL4 (HER2+ PIK3CA-mutant breast cancer xenograft), JIMT1 (HER2-low PIK3CA-mutant breast cancer xenograft), AN3CA (HER2-low PTEN-mutant endometrial cancer xenograft), and SKOV3 (HER2+ PIK3CA-mutant ovarian cancer xenograft; Fig. 5; Supplementary Tables S5 and S6). The combination of T-DXd (10 mg/kg) and capivasertib (130 mg/kg twice daily; 4 days on, 3 days off) was compared with each monotherapy and also with the combinations with trastuzumab and the clinical TOP1 inhibitor irinotecan (Fig. 5; Supplementary Figs. S11 and S12). The combination treatments were tolerated in mice (Supplementary Fig. S13). These doses of T-DXd and capivasertib were chosen as the preclinical pharmacokinetic profiles at these doses reflect the clinical exposure profiles of both drugs (Supplementary Figs. S14 and 15; refs. 37–39).
Figure 5.
Capivasertib plus T-DXd combination activity in breast, ovarian, and endometrial xenograft models. A, KPL4 HER2+ breast cancer, B, JIMT-1 HER2-low breast cancer, C, AN3CA HER2-low endometrial cancer, and D, SKOV3 HER2+ ovarian cancer xenograft models were treated with T-DXd (3 and 10 mg/kg once every 3 weeks, i.v.) and capivasertib (130 mg/kg twice daily, 4 days on/3 days off, orally). Trastuzumab (herceptin) was administered twice weekly at 4 mg/kg in KPL4 and JIMT1 and 3 mg/kg in AN3CA and SKOV3. Plots show average TV (mm3) ±SEM. Statistical comparisons and individual animal plots are shown in Supplementary Table S5. Dosing schedules are shown in more detail in Supplementary Table S6.
In KPL4 xenografts, the combination of capivasertib and T-DXd enhanced depth and durability of response over monotherapy T-DXd or capivasertib, in which capivasertib was dosed at the clinically relevant 4 days on, 3 days off schedule. T-DXd monotherapy treatment gave a TGI of 76%, capivasertib monotherapy treatment a 25% TGI, and the combination treatment a83% TGI on day 28 after treatment. Upon regrowth of the T-DXd monotherapy arm, the combination treatment demonstrated enhanced durability with a stronger response at day 46 (Fig. 5A; Supplementary Fig. S11A). The capivasertib T-DXd combination also demonstrated superior activity compared with a capivasertib–trastuzumab combination which gave a TGI of 49% on day 28 with a clinically relevant dose of trastuzumab (4 mg/kg; Fig. 5A). In the JIMT1 xenograft model, the capivasertib–T-DXd combination demonstrated enhanced antitumor benefit (84% TGI at day 24) when compared with capivasertib monotherapy (TGI 62%) and improved depth of antitumor response when compared with T-DXd monotherapy (77% TGI). Similar to the KPL4 model, the combination of capivasertib plus T-DXd was superior to capivasertib plus trastuzumab, activity of which was not much greater than capivasertib or trastuzumab monotherapies alone (67% TGI; Fig. 5B; Supplementary Fig. S11B). In both the KPL4 and JIMT1 models, the tumors were followed over time. In both models, the combination maintained the greatest tumor control, with more durable benefit relative to the best monotherapy response or the capivasertib–trastuzumab combination.
The combination benefit extended to models of other tumor types. In the HER2-low endometrial cancer xenograft model AN3CA, even saturating doses of trastuzumab (30 mg/kg) had no monotherapy activity. The combination of T-DXd (10 mg/kg) with capivasertib was the most effective with a TGI of 89% versus 62% (T-DXd) and 55% (capivasertib) in the monotherapy treatments (Fig. 5C; Supplementary Fig. S11C; Supplementary Table S5). In the ovarian HER2+ SKOV3 xenograft model, the capivasertib T-DXd combination demonstrated improved antitumor benefit when compared with either monotherapy alone (71% TGI with combination versus 45% TGI T-DXd and 44% TGI capivasertib monotherapy treatments). Interestingly, unlike the other models, the combination of capivasertib plus trastuzumab (79% TGI) was the most active treatment and provided nearly equivalent antitumor activity with trastuzumab dosed at saturating doses (30 mpk), whereas T-Dxd was dosed at a clinically relevant preclinical dose of 10 mpk (Fig. 5D; Supplementary Fig. S11D). The in vivo efficacy was largely in line with the in vitro effects of the capivasertib T-DXd combination on cell viability in these cell lines (Supplementary Fig. S16). The capivasertib–T-DXd combination increased cell death in KPL4 cells and reduced growth in AN3CA and SKOV3 cells. However, in JIMT-1 cells, combination benefit was not observed in vitro in despite combination efficacy observed in vivo, perhaps indicating that benefit is derived from inhibition of HER2. The combination was also evaluated with a lower dose of T-DXd (3 mg/kg) evaluated in SKOV3 and AN3CA xenograft models. In AN3CA, the combination benefit was comparable with that seen with 10 mg/kg (80% TGI vs. 89% TGI), whereas in SKOV3, it was less pronounced (56% TGI vs. 71% TGI; Supplementary Fig. S17; Supplementary Table S5), indicating some element of dose dependency for the effect. Collectively, the data demonstrate that the combination is effective in models representing HER2+ and HER2-low breast, ovarian, and endometrial cancers, enhancing both the depth and durability of antitumor response compared with monotherapy control treatments.
Discussion
Here, we describe the potential for the topoisomerase inhibitor–based ADC T-DXd to combine with the AKT inhibitor capivasertib. Combining T-DXd and capivasertib resulted in enhanced benefit in vitro in cell lines representative of a broad subset of tumors. In particular, combination activity was seen in breast, endometrial, gastric, and ovarian cancer cell lines. The combination benefit translated in vivo with coadministration of T-DXd and capivasertib enhancing antitumor effects in Her2+ breast cancer and ovarian cancer cell line xenograft models.
As T-DXd is comprised of trastuzumab conjugated to a topoisomerase payload, it retains both anti-HER2 potential in addition to delivering chemotherapy to target expressing cells. This means that capivasertib could potentially interact with T-DXd through multiple biological mechanisms. For example, prior work has shown capivasertib to have potential to combine with trastuzumab (6), although the combination of capivasertib with topoisomerase inhibitors has not been previously explored. Consistent with the previous studies, our study also found that capivasertib showed increased combination benefit with trastuzumab. In addition, capivasertib also showed direct combination benefit with an analog of the T-DXd payload exatecan. Importantly, despite these combination signals with trastuzumab and exatecan, neither combination was as strong or broadly active in the panel as T-DXd. This highlights the enhanced combination benefit with T-DXd compared with the individual components of the molecule, trastuzumab and the top1 inhibitor payload. Furthermore, although these data demonstrate that both components of T-DXd may contribute to the broad combination benefit observed, they also establish that capivasertib may have potential to combine with other ADCs delivering a topoisomerase inhibitor warhead.
Mechanistically, T-DXd and capivasertib seem to work independently to enhance antiproliferative effect of monotherapies by co-targeting the cell cycle. First, there was no significant increase in PI3K–AKT or DDR pathway markers observed when comparing the combination treatment with either single-agent treatment. T-DXd treatment induces DNA damage, increasing gH2AX, ATR, ATM, and Chk1/2 phosphorylation; however, lack of further enhancement of the DNA damage following the combination therapy suggests that the combination does not increase cell damage. Conversely, no detectable enhancement in AKT pathway biomarkers was observed following T-DXd monotherapy treatment, and the combination treatment did not enhance AKT pathway inhibition when compared with capivasertib monotherapy treatment. The data suggest that the enhanced activity of the combination may be associated with T-DXd and capivasertib independently targeting complementary pathways that converge downstream leading to cell-cycle arrest and increased cell death. This mechanistic hypothesis is consistent with the pharmacologic effects observed by the combination and aligns with known effects in which impaired G2/M cell-cycle arrest, increasing the potential for cell death. Despite the mixed effects on cell-cycle biomarkers and the FACS-based cell cycle profiles, increased cell death was observed independent of the point of impact in the cell cycle, suggesting that cell death may be occurring at multiple points in the cell cycle. A caveat of this current study is that a specific mechanism driving the increased cell death and reduced proliferation was not identified; further work will be required to identify the specific mechanisms involved. These could include modulation of AKT-mediated apoptosis checkpoints, restriction in nucleotide availability, or increase in replication stress due to slowing of cell-cycle completion (40).
More broadly, there are reports that PI3K/AKT activation is frequent across multiple cancers, including ovarian cancer, endometrial cancer, and all subtypes of breast cancer. Aberrant PI3K/AKT activation can influence efflux pump expression, particularly ABC transporters which can influence topoisomerase I inhibitors, including irinotecan and SN-38 (41–43). It is most likely that a number of mechanisms can contribute to the combination benefit depending on context.
Several mechanisms have been associated with resistance to T-DXd, including mechanisms related to HER2 expression, payload, drug efflux pumps, and alternative signaling pathways (44). Reductions in HER2 expression or ERBB2 mutations are major T-DXd resistance mechanisms to in both clinical samples and preclinical models, with decreased or complete loss of HER2 expression in nearly half of samples (45-47). Analysis of on progression ctDNA from T-DXd–treated patients with breast cancer revealed that increased PIK3CA alterations were the most commonly emerging mutation (45). Loss-of-function mutations in PTEN alterations are also associated with worse progression-free survival and poor response to T-DXd but not to the TROP-2–targeting sacituzumab govitecan (46). Together, these data suggest that PI3K pathway alterations are likely involved in acquired resistance to T-DXd; however, further work is needed to determine whether T-DXd combined with PI3K–AKT inhibitors can affect acquired resistance.
The combination of T-DXd with capivasertib was shown to be superior to the trastuzumab plus capivasertib combination in HER2+ and HER2-low breast and endometrial xenograft models. Importantly, administration of T-DXd and capivasertib using clinically relevant dose and schedules was able to deliver enhanced antitumor benefit without increasing body weight loss over single-agent controls. However given the effects on the cell cycle, it is possible that exploration of alternative schedules of dosing two drugs may further enhance their efficacy, as has been seen in other combination approaches (5), and be used to assess the effect on induction of pharmacodynamic biomarkers over time in vivo. Potential tolerability issues were explored further using 2D in vitro bone marrow assays, which demonstrated low risk of exacerbated hematologic toxicity. This, coupled with the enhanced combination benefit only observed in subsets of tumor cell lines, suggests that this combination does not have a broad impact on proliferating cells and gives further confidence that the effects are not due to general cytotoxicity. However, further work would be required to determine whether the combination has additive effects in other normal tissues.
Given the targeted nature of both T-DXd and capivasertib, it was important to explore whether the combination response is enriched in relevant molecular segments, such as HER2 overexpression or PI3K–AKT pathway–altered models, to enable more accurate translation to the clinic. T-DXd has demonstrated preclinical and clinical efficacy in both HER2-high– and HER2-low–expressing tumors. It is approved by the FDA for the treatment of HER2+ and HER2-low breast cancer, HER2-mutant non–small cell lung cancer, and HER2+ gastric cancer (14, 15, 18). Aligned with these clinical data, combination benefit was evident in HER2-high and HER2-low preclinical models in both the in vitro and in vivo experiments. Across the 300 cell lines and in vivo, the combination of T-DXd plus capivasertib showed benefit in both HER2-high and HER2-low expressing models. Therefore, combination of T-DXd with capivasertib has the potential to benefit patients across a broad range of HER2 expression levels.
Monotherapy response to capivasertib is seen in cell lines with and without alterations in PIK3CA, PTEN, and AKT-1, although there is some enrichment in sensitivity in altered cell lines (6, 36). In this study, no statistically significant association was observed between the combination benefit of capivasertib plus T-DXd or exatecan and PI3K–AKT pathway alterations across 300 cell lines treated with the combination or in a bespoke breast cancer cell line panel. These observations build on the broad combination potential for capivasertib with fulvestrant, taxane-based chemotherapy, and even BLC2 antagonists observed in PI3K–AKT pathway–altered and nonaltered models (4, 5, 48).
Overall, the findings from this study suggest that combining T-DXd with capivasertib has the potential to be an effective treatment for HER2+ and HER2-low tumors. Although the combination activity may be enriched in tumors with PIK3CA or PTEN mutations, efficacy is also observed in tumors without PI3K–AKT pathway alterations. T-DXd in combination with capivasertib is currently being investigated in a phase 1b clinical study in patients with HER2-low advanced or metastatic breast cancer (DESTINY-Breast08, NCT04556773). The results of this trial were recently reported at SABCS 2024. In agreement with the findings in this manuscript, T-DXd in combination with capivasertib demonstrated preliminary antitumor activity, with no unexpected side effects (49).
In summary, T-DXd has potential to combine with the AKT inhibitor capivasertib to enhance cytotoxic activity in a number of disease settings, including breast, gastric, endometrial, and gastric cancers independent of both Her2 overexpression or PI3K–AKT alterations.
Supplementary Material
Supplementary Table 1. Cell culture details
Supplementary Table 2. Antibody details
Supplementary Table 3. Summary of data from selected breast and gastric cell panel in vitro screen with capivasertib, T-DXd and combination treatment.
Supplementary Table 4. Summary of data from selected breast and gastric cell panel in vitro screen for capivasertib, exatecan, trastuzumab (Herceptin) and combination treatment.
Supplementary Table 5. Summary of TGI for all in vivo studies
Supplementary Table 6. Study design for in vivo efficacy evaluation
Supplementary Figure 1. Capivasertib plus T-DXd combination treatment activity in breast and gastric cell lines.
Supplementary Figure 2. Cell viability values at clinically relevant concentrations of capivasertib and T-DXd.
Supplementary Figure 3. In vitro capivasertib plus exatecan combination benefit in cell lines.
Supplementary Figure 4. Capivasertib plus exatecan combination activity in breast and gastric cell lines.
Supplementary Figure 5. Comparison of Loewe, Bliss and HSA scores for the capivasertib T-DXd combination in human primary bone marrow-derived CD34+ cells differentiated into erythroid, myeloid or megakaryocyte lineages
Supplementary Figure 6. Western blot total-protein markers in HER2-low (AN3CA, EFM19) and HER2-positive (KPL4) cell lines treated with capivasertib alone or in combination with T-DXd.
Supplementary Figure 7. Western blot analysis of NCI-N87 cells treated with capivasertib, T-DXd monotherapy or the combination.
Supplementary Figure 8. Active cell cycle distribution analysis of KPL4, EFM19 and KPL4 cells treated with T-DXd or capivasertib monotherapy, or the combination
Supplementary Figure 9. HSA distribution across all cell lines (solid and haematological) in a broad cell panel treated with capivasertib in combination with T-DXd or exatecan for 6 days.
Supplementary Figure 10. Association between benefit from the capivasertib plus T-DXd combination and PI3K-AKT pathway alterations in gastric, endometrial and uterine cell lines
Supplementary Figure 11. Individual animals plots for breast, ovarian and endometrial xenograft models treated with T-DXd, capivasertib or the combination.
Supplementary Figure 12. Capivasertib and irinotecan monotherapy, and the combination activity in tumour xenograft models.
Supplementary Figure 13. Body weight of animals in anti-tumour studies.
Supplementary Figure 14. Aligning mouse and human pharmacokinetic profiles for T-DXd
Supplementary Figure 15. Capivasertib mouse PK parameters
Supplementary Figure 16. In vitro activity of capivasertib plus T-DXd in cell lines used for in vivo studies
Supplementary Figure 17. In vivo efficacy of capivasertib plus T-DXd in AN3CA and SKOV3
Acknowledgments
We would like to thank the in vivo group at AstraZeneca Gatehouse Park Boston for support with in vivo studies. Capivasertib (AZD5363) was discovered by AstraZeneca subsequent to a collaboration with Astex Therapeutics (and its collaboration with the Institute of Cancer Research and Cancer Research Technology Limited). We would like to thank Michael Caprio for cell provision for the focused cell panel screen and Deepa Bhavsar for supporting the focused cell panel screen.
Footnotes
Note: Supplementary data for this article are available at Molecular Cancer Therapeutics Online (http://mct.aacrjournals.org/).
Contributor Information
Jerome T. Mettetal, Email: jay.mettetal@astrazeneca.com.
Simon T. Barry, Email: simon.t.barry@astrazeneca.com.
Data Availability
All raw data underlying this article’s findings are available upon request to the corresponding author.
Authors’ Disclosures
A.C. Bashi reports other support from AstraZeneca outside the submitted work. T.A. Proia reports other support from AstraZeneca outside the submitted work. A. Nelson is an AstraZeneca employee and shareholder. S.J. Randle is an employee and shareholder at AstraZeneca. S. Agrawal reports other support from AstraZeneca during the conduct of the study. D. Carroll reports other support from AstraZeneca during the conduct of the study and outside the submitted work. J.T. Mettetal is an AstraZeneca employee and shareholder. S.T. Barry is an AstraZeneca employee and shareholder. No disclosures were reported by the other authors.
Authors’ Contributions
A.C. Bashi: Conceptualization, data curation, formal analysis, supervision, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing, study design, data generation, data interpretation. T.A. Proia: Formal analysis, supervision, investigation, visualization, methodology, writing–review and editing. M. Lawson: Formal analysis, visualization, methodology, writing–review and editing, data generation. A. Nelson: Formal analysis, visualization, writing–review and editing. L. Ireland: Formal analysis, visualization, methodology, writing–review and editing, data generation. S.J. Randle: Formal analysis, methodology, writing–review and editing, data generation. S. Agrawal: Formal analysis, visualization, writing–review and editing. A. Rosen: Methodology, writing–review and editing, data generation. D. Carroll: Writing–review and editing, data interpretation. J.T. Mettetal: Conceptualization, supervision, writing–original draft, writing–review and editing, data interpretation. S.T. Barry: Conceptualization, supervision, writing–original draft, writing–review and editing, data interpretation.
References
- 1. He Y, Sun MM, Zhang GG, Yang J, Chen KS, Xu WW, et al. Targeting PI3K/Akt signal transduction for cancer therapy. Signal Transduct Target Ther 2021;6:425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Manning BD, Toker A. AKT/PKB signaling: navigating the network. Cell 2017;169:381–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Hua H, Zhang H, Chen J, Wang J, Liu J, Jiang Y. Targeting Akt in cancer for precision therapy. J Hematol Oncol 2021;14:128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hopcroft L, Wigmore EM, Williamson SC, Ros S, Eberlein C, Moss JI, et al. Combining the AKT inhibitor capivasertib and SERD fulvestrant is effective in palbociclib-resistant ER+ breast cancer preclinical models. NPJ Breast Cancer 2023;9:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Eberlein C, Williamson SC, Hopcroft L, Ros S, Moss JI, Kerr J, et al. Capivasertib combines with docetaxel to enhance anti-tumour activity through inhibition of AKT-mediated survival mechanisms in prostate cancer. Br J Cancer 2024;130:1377–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Davies BR, Greenwood H, Dudley P, Crafter C, Yu DH, Zhang J, et al. Preclinical pharmacology of AZD5363, an inhibitor of AKT: pharmacodynamics, antitumor activity, and correlation of monotherapy activity with genetic background. Mol Cancer Ther 2012;11:873–87. [DOI] [PubMed] [Google Scholar]
- 7. Turner NC, Oliveira M, Howell SJ, Dalenc F, Cortes J, Gomez Moreno HL, et al. Capivasertib in hormone receptor-positive advanced breast cancer. N Engl J Med 2023;388:2058–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lawson M, Cureton N, Ros S, Cheraghchi-Bashi A, Urosevic J, D’Arcy S, et al. The next-generation oral selective estrogen receptor degrader camizestrant (AZD9833) suppresses ER+ breast cancer growth and overcomes endocrine and CDK4/6 inhibitor resistance. Cancer Res 2023;83:3989–4004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Crabb SJ, Griffiths G, Marwood E, Dunkley D, Downs N, Martin K, et al. Pan-AKT inhibitor capivasertib with docetaxel and prednisolone in metastatic castration-resistant prostate cancer: a randomized, placebo-controlled phase II trial (ProCAID). J Clin Oncol 2021;39:190–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. De Velasco MA, Kura Y, Yoshikawa K, Nishio K, Davies BR, Uemura H. Efficacy of targeted AKT inhibition in genetically engineered mouse models of PTEN-deficient prostate cancer. Oncotarget 2016;7:15959–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Marques RB, Aghai A, de Ridder CMA, Stuurman D, Hoeben S, Boer A, et al. High efficacy of combination therapy using PI3K/AKT inhibitors with androgen deprivation in prostate cancer preclinical models. Eur Urol 2015;67:1177–85. [DOI] [PubMed] [Google Scholar]
- 12. Toren P, Kim S, Cordonnier T, Crafter C, Davies BR, Fazli L, et al. Combination AZD5363 with enzalutamide significantly delays enzalutamide-resistant prostate cancer in preclinical models. Eur Urol 2015;67:986–90. [DOI] [PubMed] [Google Scholar]
- 13. Searle EJ, Telfer BA, Mukherjee D, Forster DM, Davies BR, Williams KJ, et al. Akt inhibition improves long-term tumour control following radiotherapy by altering the microenvironment. EMBO Mol Med 2017;9:1646–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Modi S, Saura C, Yamashita T, Park YH, Kim SB, Tamura K, et al. Trastuzumab deruxtecan in previously treated HER2-positive breast cancer. N Engl J Med 2020;382:610–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Modi S, Park H, Murthy RK, Iwata H, Tamura K, Tsurutani J, et al. Antitumor activity and safety of trastuzumab deruxtecan in patients with HER2-low-expressing advanced breast cancer: results from a phase Ib study. J Clin Oncol 2020;38:1887–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Modi S, Gambhire D, Cameron D. Trastuzumab deruxtecan in HER2-low breast cancer. Reply. N Engl J Med 2022;387:1145–6. [DOI] [PubMed] [Google Scholar]
- 17. Goto K, Goto Y, Kubo T, Ninomiya K, Kim SW, Planchard D, et al. Trastuzumab deruxtecan in patients with HER2-mutant metastatic non-small-cell lung cancer: primary results from the randomized, phase II DESTINY-lung02 trial. J Clin Oncol 2023;41:4852–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Shitara K, Bang YJ, Iwasa S, Sugimoto N, Ryu MH, Sakai D, et al. Trastuzumab deruxtecan in previously treated HER2-positive gastric cancer. N Engl J Med 2020;382:2419–30. [DOI] [PubMed] [Google Scholar]
- 19. Meric-Bernstam F, Makker V, Oaknin A, Oh DY, Banerjee S, González-Martín A, et al. Efficacy and safety of trastuzumab deruxtecan in patients with HER2-expressing solid tumors: primary results from the DESTINY-PanTumor02 phase II trial. J Clin Oncol 2024;42:47–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Swain SM, Shastry M, Hamilton E. Targeting HER2-positive breast cancer: advances and future directions. Nat Rev Drug Discov 2023;22:101–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Fujimoto Y, Morita TY, Ohashi A, Haeno H, Hakozaki Y, Fujii M, et al. Combination treatment with a PI3K/Akt/mTOR pathway inhibitor overcomes resistance to anti-HER2 therapy in PIK3CA-mutant HER2-positive breast cancer cells. Sci Rep 2020;10:21762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Crafter C, Vincent JP, Tang E, Dudley P, James NH, Klinowska T, et al. Combining AZD8931, a novel EGFR/HER2/HER3 signalling inhibitor, with AZD5363 limits AKT inhibitor induced feedback and enhances antitumour efficacy in HER2-amplified breast cancer models. Int J Oncol 2015;47:446–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Ogitani Y, Aida T, Hagihara K, Yamaguchi J, Ishii C, Harada N, et al. DS-8201a, A novel HER2-targeting adc with a novel DNA topoisomerase I inhibitor, demonstrates a promising antitumor efficacy with differentiation from T-DM1. Clin Cancer Res 2016;22:5097–108. [DOI] [PubMed] [Google Scholar]
- 24. Rasti AR, Guimaraes-Young A, Datko F, Borges VF, Aisner DL, Shagisultanova E. PIK3CA mutations drive therapeutic resistance in human epidermal growth factor receptor 2-positive breast cancer. JCO Precis Oncol 2022;6:e2100370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Loibl S, Majewski I, Guarneri V, Nekljudova V, Holmes E, Bria E, et al. PIK3CA mutations are associated with reduced pathological complete response rates in primary HER2-positive breast cancer: pooled analysis of 967 patients from five prospective trials investigating lapatinib and trastuzumab. Ann Oncol 2016;27:1519–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Tolaney SM, Jiang Z, Zhang Q, Barroso-Sousa R, Park YH, Rimawi MF, et al. Trastuzumab deruxtecan plus pertuzumab for HER2-positive metastatic breast cancer. N Engl J Med 2026;394:551–62. [DOI] [PubMed] [Google Scholar]
- 27. Liu R, Chen Y, Liu G, Li C, Song Y, Cao Z, et al. PI3K/AKT pathway as a key link modulates the multidrug resistance of cancers. Cell Death Dis 2020;11:797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Orlando BJ, Liao M. ABCG2 transports anticancer drugs via a closed-to-open switch. Nat Commun 2020;11:2264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Berns K, Horlings HM, Hennessy BT, Madiredjo M, Hijmans EM, Beelen K, et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer Cell 2007;12:395–402. [DOI] [PubMed] [Google Scholar]
- 30. Foucquier J, Guedj M. Analysis of drug combinations: current methodological landscape. Pharmacol Res Perspect 2015;3:e00149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Yadav B, Wennerberg K, Aittokallio T, Tang J. Searching for drug synergy in complex dose-response landscapes using an interaction potency model. Comput Struct Biotechnol J 2015;13:504–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Hurvitz SA, Hegg R, Chung WP, Im SA, Jacot W, Ganju V, et al. Trastuzumab deruxtecan versus trastuzumab emtansine in patients with HER2-positive metastatic breast cancer: updated results from DESTINY-Breast03, a randomised, open-label, phase 3 trial. Lancet 2023;401:105–17. [DOI] [PubMed] [Google Scholar]
- 33. Huang TH, Chen HC, Chou SM, Yang YC, Fan JR, Li TK. Cellular processing determinants for the activation of damage signals in response to topoisomerase I-linked DNA breakage. Cell Res 2010;20:1060–75. [DOI] [PubMed] [Google Scholar]
- 34. Kurebayashi J, Otsuki T, Tang CK, Kurosumi M, Yamamoto S, Tanaka K, et al. Isolation and characterization of a new human breast cancer cell line, KPL-4, expressing the Erb B family receptors and interleukin-6. Br J Cancer 1999;79:707–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Cowley GS, Weir BA, Vazquez F, Tamayo P, Scott JA, Rusin S, et al. Parallel genome-scale loss of function screens in 216 cancer cell lines for the identification of context-specific genetic dependencies. Sci Data 2014;1:140035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Bashi AC, Coker EA, Bulusu KC, Jaaks P, Crafter C, Lightfoot H, et al. Large-scale pan-cancer cell line screening identifies actionable and effective drug combinations. Cancer Discov 2024;14:846–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Vasalou C, Proia TA, Kazlauskas L, Przybyla A, Sung M, Mamidi S, et al. Quantitative evaluation of trastuzumab deruxtecan pharmacokinetics and pharmacodynamics in mouse models of varying degrees of HER2 expression. CPT Pharmacometrics Syst Pharmacol 2024;13:994–1005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Yin O, Xiong Y, Endo S, Yoshihara K, Garimella T, AbuTarif M, et al. Population pharmacokinetics of trastuzumab deruxtecan in patients with HER2-positive breast cancer and other solid tumors. Clin Pharmacol Ther 2021;109:1314–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Fernandez-Teruel C, Cullberg M, Eberlein C, Barry ST, Zhou D. Population pharmacokinetics of capivasertib in patients with advanced or metastatic solid tumours. Clin Pharmacokinet 2024;63:1191–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Sementino E, Hassan D, Bellacosa A, Testa JR. AKT and the hallmarks of cancer. Cancer Res 2024;84:4126–39. [DOI] [PubMed] [Google Scholar]
- 41. Tazzari PL, Cappellini A, Ricci F, Evangelisti C, Papa V, Grafone T, et al. Multidrug resistance-associated protein 1 expression is under the control of the phosphoinositide 3 kinase/Akt signal transduction network in human acute myelogenous leukemia blasts. Leukemia 2007;21:427–38. [DOI] [PubMed] [Google Scholar]
- 42. Gu J, Huang W, Wang X, Zhang J, Tao T, Zheng Y, et al. Hsa-miR-3178/RhoB/PI3K/Akt, a novel signaling pathway regulates ABC transporters to reverse gemcitabine resistance in pancreatic cancer. Mol Cancer 2022;21:112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Abdul-Ghani R, Serra V, Györffy B, Jürchott K, Solf A, Dietel M, et al. The PI3K inhibitor LY294002 blocks drug export from resistant colon carcinoma cells overexpressing MRP1. Oncogene 2006;25:1743–52. [DOI] [PubMed] [Google Scholar]
- 44. Cerci B, Saatci O, Basik M, Sahin O. Mechanisms of resistance to antibody-drug conjugates in breast cancer. Drug Resist Updat 2026;85:101353. [DOI] [PubMed] [Google Scholar]
- 45. Chang H, Anawate I, Low A, Huang S-P, Maues J, Hodgdon C, et al. Abstract PO4-14-10: circulating tumor DNA as a biomarker for ADCs in metastatic breast cancer. Cancer Res 2024;84(9_Suppl):PO4-14-10. [Google Scholar]
- 46. Mai N, Klar Lieberman MM, Ferraro E, Bromberg M, Chen Y, Razavi P, et al. Sequential antibody-drug conjugate therapy in patients with metastatic breast cancer treated with sacituzumab govitecan and trastuzumab deruxtecan. JCO Precis Oncol 2025;9:e2400898. [DOI] [PubMed] [Google Scholar]
- 47. Mosele F, Deluche E, Lusque A, Le Bescond L, Filleron T, Pradat Y, et al. Trastuzumab deruxtecan in metastatic breast cancer with variable HER2 expression: the phase 2 DAISY trial. Nat Med 2023;29:2110–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Dunn S, Eberlein C, Yu J, Gris-Oliver A, Ong SH, Yelland U, et al. AKT-mTORC1 reactivation is the dominant resistance driver for PI3Kbeta/AKT inhibitors in PTEN-null breast cancer and can be overcome by combining with Mcl-1 inhibitors. Oncogene 2022;41:5046–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Jhaveri K, André F, Hamilton E, Schmid P, Anders CK, Wildiers H, et al. Trastuzumab deruxtecan (T-DXd) in combination with capecitabine or capivasertib in patients with HER2-low metastatic breast cancer: a Phase 1b, multicenter, open-label study (DESTINY-Breast08). Clin Cancer Res 2025;31(12_Suppl):P3-09-17. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1. Cell culture details
Supplementary Table 2. Antibody details
Supplementary Table 3. Summary of data from selected breast and gastric cell panel in vitro screen with capivasertib, T-DXd and combination treatment.
Supplementary Table 4. Summary of data from selected breast and gastric cell panel in vitro screen for capivasertib, exatecan, trastuzumab (Herceptin) and combination treatment.
Supplementary Table 5. Summary of TGI for all in vivo studies
Supplementary Table 6. Study design for in vivo efficacy evaluation
Supplementary Figure 1. Capivasertib plus T-DXd combination treatment activity in breast and gastric cell lines.
Supplementary Figure 2. Cell viability values at clinically relevant concentrations of capivasertib and T-DXd.
Supplementary Figure 3. In vitro capivasertib plus exatecan combination benefit in cell lines.
Supplementary Figure 4. Capivasertib plus exatecan combination activity in breast and gastric cell lines.
Supplementary Figure 5. Comparison of Loewe, Bliss and HSA scores for the capivasertib T-DXd combination in human primary bone marrow-derived CD34+ cells differentiated into erythroid, myeloid or megakaryocyte lineages
Supplementary Figure 6. Western blot total-protein markers in HER2-low (AN3CA, EFM19) and HER2-positive (KPL4) cell lines treated with capivasertib alone or in combination with T-DXd.
Supplementary Figure 7. Western blot analysis of NCI-N87 cells treated with capivasertib, T-DXd monotherapy or the combination.
Supplementary Figure 8. Active cell cycle distribution analysis of KPL4, EFM19 and KPL4 cells treated with T-DXd or capivasertib monotherapy, or the combination
Supplementary Figure 9. HSA distribution across all cell lines (solid and haematological) in a broad cell panel treated with capivasertib in combination with T-DXd or exatecan for 6 days.
Supplementary Figure 10. Association between benefit from the capivasertib plus T-DXd combination and PI3K-AKT pathway alterations in gastric, endometrial and uterine cell lines
Supplementary Figure 11. Individual animals plots for breast, ovarian and endometrial xenograft models treated with T-DXd, capivasertib or the combination.
Supplementary Figure 12. Capivasertib and irinotecan monotherapy, and the combination activity in tumour xenograft models.
Supplementary Figure 13. Body weight of animals in anti-tumour studies.
Supplementary Figure 14. Aligning mouse and human pharmacokinetic profiles for T-DXd
Supplementary Figure 15. Capivasertib mouse PK parameters
Supplementary Figure 16. In vitro activity of capivasertib plus T-DXd in cell lines used for in vivo studies
Supplementary Figure 17. In vivo efficacy of capivasertib plus T-DXd in AN3CA and SKOV3
Data Availability Statement
All raw data underlying this article’s findings are available upon request to the corresponding author.