Targeted Sequencing of Tubo-ovarian and Peritoneal High-grade Serous Carcinoma With Wild-type p53 Immunostaining Pattern

Abstract

Background/Aim: We aimed to demonstrate the use of next-generation sequencing (NGS) to confirm the presence of tumor protein 53 (TP53) mutations in tubo-ovarian and peritoneal high-grade serous carcinoma (HGSC) with a wild-type p53 immunostaining pattern and investigate whether the TP53 mutational status is altered by chemotherapy. Materials and Methods: A commercial NGS panel comprising 171 genes was used to analyze the genetic profiles of 15 HGSC samples. Paired specimens obtained before and after chemotherapy were available for four patients. Results: All examined samples exhibited TP53 mutations. For all the patients who underwent neoadjuvant or postoperative adjuvant chemotherapy, TP53 mutations identified in samples obtained after chemotherapy were the same as those detected in pre-chemotherapeutic samples. Conclusion: HGSCs exhibit TP53 mutations even though a subset of HGSCs displayed a wild-type p53 immunostaining pattern. Chemotherapy does not affect the TP53 mutational status in HGSC.

Tumor protein 53 (TP53) is the most frequently mutated gene in malignancies. TP53 encodes for a transcription factor, p53, that initiates the transcription of genes involved in cell cycle arrest, cellular senescence, apoptosis, metabolism, DNA repair, and other processes following cellular stress (1,2). p53, a critical tumor suppressor, plays a fundamental and multifaceted role in the development and progression of malignancies (3). In the absence of cellular stress, wild-type p53 is maintained at low levels, while in response to cellular stress, p53 is stabilized and activated through numerous mechanisms (4,5). Activated wild-type p53 promotes processes consistent with tumor suppression, whereas mutation of p53 results in the loss of these tumor-suppressive functions.

Wild-type p53 is relatively unstable and has a short half-life, which makes it undetectable by immunostaining (6,7). In contrast, mutant p53 has a much longer half-life and accumulates in the nucleus, thereby is detected by immunohistochemistry (7). TP53 mutations include single-base substitutions leading to missense or non-sense point mutations, in-frame deletions or insertions, frameshift deletions or insertions, as well as mutations that affect splicing sites (5). Diffuse and strong nuclear p53 expression is regarded as indicative of a missense TP53 mutation (8,9), and the complete absence of p53 immunoreactivity results from a nonsense TP53 mutation, leading to the formation of a truncated, non-immunoreactive protein (10-12).

Tubo-ovarian and peritoneal high-grade serous carcinoma (HGSC) is characterized by high frequency of pathogenic TP53 mutations. HGSC is the eighth-most frequent cause of cancer-related deaths in women worldwide (13). Most ovarian carcinoma cases are diagnosed at advanced stages, at which point, the five-year survival rate is approximately 25% (5,14). The existing therapeutic options for patients with tubo-ovarian and peritoneal HGSC are limited to aggressive debulking surgery and postoperative platinum-based adjuvant chemotherapy. An increased understanding of the alterations in the expression of genes and proteins involved in ovarian carcinogenesis may aid in improving the diagnosis and treatment of HGSC. A diagnostic or prognostic biomarker for ovarian carcinoma is, thus, urgently needed to guide the treatment of these patients.

It has recently been suggested that all HGSC cases are, in fact, TP53-mutants (15). This finding is in contrast to that previously known from studies that used less sensitive methods such as direct sequencing and focused on hot-spot regions of this gene only (7,16,17). A precise validation of this practice by comparison with detailed sequencing data has been limited. We recently reported the direct sequencing results of TP53 in a cohort of HGSC in parallel with p53 immunostaining results (9). In this study, next-generation sequencing (NGS) was used to confirm the data previously obtained and to investigate whether the TP53 mutational status is altered by preoperative neoadjuvant or postoperative adjuvant chemotherapy in a larger cohort of HGSC with wild-type p53 immunostaining pattern.

Materials and Methods

Case selection. Following approval (4-2017-0993) by the Institutional Review Board, 240 cases of tubo-ovarian and peritoneal HGSC were initially selected from the archives of the Department of Pathology at the Severance Hospital (Seoul, Republic of Korea). We extracted 11 HGSC cases showing wild-type p53 immunostaining pattern. Clinical and pathological information, including the age of patient at initial diagnosis, tumor location, histological grade, International Federation of Gynecology and Obstetrics (FIGO) stage, type of surgical treatment, and addition of neoadjuvant or postoperative adjuvant chemotherapy, was obtained from the electronic medical record system and pathology reports. Two patients whose tumors showed mutant p53 immunostaining pattern (as a positive control) and one patient who underwent hysterectomy with left salpingo-oophorectomy for uterine leiomyoma (as a negative control) were also included.

Pathological examination. The resected tissues were initially examined by two pathologists, followed by fixation in 10% neutral-buffered formalin for 12-24 h. The tissues were then examined macroscopically and sectioned. After processing with an automatic tissue processor (Peloris II, Leica Microsystems, Newcastle Upon Tyne, UK), the sections were embedded in paraffin blocks. Four-micrometer-thick slices were sectioned from each formalin-fixed, paraffin-embedded (FFPE) tissue block using a rotary microtome (RM2245, Leica Microsystems) and stained with hematoxylin and eosin using an automatic staining instrument (Ventana Symphony System, Ventana Medical Systems, Tucson, AZ, USA). After staining, the slides were covered with a glass coverslip and sent to a board-certified pathologist specialized in gynecological oncology. The pathologist examined the hematoxylin and eosin-stained slides by light microscopy (BX43 System Microscope, Olympus, Tokyo, Japan) and made pathological diagnoses. In addition, the most representative slide for each case was chosen for subsequent immunostaining and sequencing.

Immunohistochemical staining. Immunostaining was performed using an automatic instrument [Ventana Benchmark XT (Ventana Medical Systems)] according to the manufacturer’s recommendations (9,18-35). Antigen retrieval was performed using Cell Conditioning Solution (CC1, Ventana Medical Systems). The 4-μm-thick, formalin-fixed, paraffin-embedded sections were incubated with anti-p53 antibody (1:300, clone DO-7, Novocastra, Newcastle Upon Tyne, UK). After chromogenic visualization using an ultraView Universal DAB Detection Kit (Ventana Medical Systems), sections were counterstained with hematoxylin. Appropriate positive and negative controls were concurrently stained to validate the staining method. Negative control was prepared by substituting non-immune serum for primary antibody, which resulted in no detectable staining. The p53 immunostaining pattern was interpreted as a missense mutation, nonsense mutation, or wild-type pattern when p53 expression was diffuse and strong (>60% of tumor cell nuclei), completely absent (0%), or focal and weakly positive, respectively (9,19,20,23).

Targeted sequencing. Genomic DNA was extracted using a QIAamp DNA FFPE Tissue Kit (Qiagen, Valencia, CA, USA). Coding exons and the flanking regions of 171 genes (Table I) were enriched using the SureSelectXT Reagent Kit (Agilent Technologies, Santa Clara, CA, USA) (19,23). The products were sequenced on a HiSeq 2500 System (Illumina, San Diego, CA, USA) using paired-end reads. The reads were aligned to a reference genome sequence (Genome Reference Consortium Human Build 37) obtained from the University of California Santa Cruz Genome Browser database (https://genome.ucsc.edu/) (36), and duplicate reads were removed. Burrows-Wheeler Aligner (http://bio-bwa.sourceforge.net/), SAMtools (http://samtools.sourceforge.net/), Picard Tools (http://broadinstitute.github.io/picard/), and Genome Analysis Toolkit (https://software.broadinstitute.org/gatk/) were used for sorting Sequence Alignment Map/Binary Alignment Map files, duplicate marking, and local realignment, respectively. Local realignment and base recalibration were performed using the Single Nucleotide Polymorphism Database (https://www.ncbi.nlm.nih.gov/ projects/SNP/), Mills indel reference (37), HapMap (https:// www.ncbi.nlm.nih.gov/probe/docs/projhapmap/), and Omni (http:// www.internationalgenome.org/category/omni/). Single nucleotide variants, insertions, and deletions were identified using the MuTect (http://archive.broadinstitute.org/cancer/cga/mutect) and Pindel (http://gmt.genome.wustl.edu/packages/pindel/), respectively. ANNOVAR (http://annovar.openbioinformatics.org/) was used to annotate the detected variants. Any single nucleotide variant present at >0.1% in the Exome Variant Server (https:// evs.gs.washington.edu/) or Single Nucleotide Polymorphism Database was filtered. The variants present in the Catalogue of Somatic Mutations in Cancer (https://cancer.sanger.ac.uk/) were reviewed.

Table I. Gene list.

Results

Clinicopathological characteristics. Table II summarizes the clinicopathological characteristics of 11 patients with HGSC showing wild-type p53 immunostaining pattern. Patient age at initial diagnosis ranged between 51 and 79 years (mean=63.7 years; median=61 years). Histological grade was 3 in eight (72.7%) patients and 2 in three (27.3%) patients. Six (54.5%) patients presented at FIGO stage IIIC and 3 (27.3%) at FIGO stage IVB. The remaining two patients presented at FIGO stage IC (1/11; 9.1%) and IIB (1/11; 9.1%), respectively. All (11/11; 100.0%) patients underwent postoperative platinum-based adjuvant chemotherapy. Paired tumor tissue samples were obtained before and after chemotherapy in two (18.2%) patients; pre-chemotherapeutic samples (samples 8-1 and 9-1) were taken from the tumor tissues obtained with primary debulking surgery (PDS), and post-chemotherapeutic samples (samples 8-2 and 9-2) were taken from the recurrent tumor tissues obtained with secondary debulking surgery (SDS). Two (18.2%) patients underwent platinum-based neoadjuvant chemotherapy followed by interval debulking surgery (IDS). Similarly, paired samples were obtained before and after chemotherapy in two patients; pre-chemotherapeutic samples (sample 10-1 and 11-1) were taken from biopsy specimens obtained with diagnostic laparoscopy, and post-chemotherapeutic samples (sample 10-2 and 11-2) were taken from the tumor tissues obtained with IDS. Eighteen tissue samples were obtained from the following sites: the right ovary (6/18; 33.3%), left ovary (2/18; 11.1%), left fallopian tube (2/18; 11.1%), pelvic peritoneum (4/18; 22.2%), and omentum (4/18; 22.2%).

Table II. Clinicopathological characteristics of patients with HGSC showing wild-type p53 immunostaining pattern.

FIGO: International Federation of Gynecology and Obstetrics; HGSC: high-grade serous carcinoma; NA: not applicable.

All cases showed characteristic histopathological features of HGSC, represented by destructive infiltration of tumor cells forming branching papillary fronds, slit-like fenestrations, and complex glandular architecture (Figure 1A). The tumor cells showed severe nuclear atypia, frequent mitoses, and atypical mitotic figures (Figure 1B). Immunostaining for p53 was performed on 18 tissue samples. In 11 HGSC and one normal fallopian tube sample, p53 expression was patchy and showed weak-to-moderate intensity in the tumor cell nuclei, similar to the wild-type pattern (Figure 1C). In the remaining two HGSC samples, diffuse and strong nuclear p53 immunoreactivity (missense mutation pattern; Figure 1D) and complete absence of p53 expression (nonsense mutation pattern; Figure 1E) were observed, respectively.

Figure 1. Histopathological and immunohistochemical findings of high-grade serous carcinoma. A: Diffuse growth of tumor cells displaying destructive stromal infiltration, solid and papillary architecture, slit-like fenestrations, and complex glands with intraluminal necrotic debris. B: Higher magnification of the area highlighted in A (blue box) reveals tumor cells showing severe nuclear pleomorphism, conspicuous nucleoli, frequent mitoses, and atypical mitotic figures. C: Patchy and weak p53 expression (wild-type pattern). D: Diffuse and strong p53 expression (missense mutation pattern). E: Complete absence of p53 expression (nonsense mutation pattern). A: Original magnification, A, 40×; B, 400×; C, D, and E, 200×.

NGS results. Table III summarizes the targeted sequencing results of 18 tissue samples, including 15 HGSC samples with wild-type p53 immunostaining pattern, two HGSC samples with mutant p53 immunostaining pattern, and one fallopian tube sample, showing wild-type staining pattern. TP53 mutations were identified in all (17/17; 100.0%) HGSC tissue samples, but not in the normal fallopian tube sample. Six (33.3%) samples showed a nonsense mutation of the TP53 gene (samples 3, 4, 5, 6, 10-1, and 10-2). Missense TP53 mutations occurred in one (5.6%) sample (sample 7). Four (22.2%) samples had frameshift mutations, two of which were frameshift deletions (samples 9-1 and 9-2) and two insertions (samples 11-1 and 11-2). Splice site mutations were identified in four (22.2%) samples (samples 1, 2, 8-1, and 8-2). Arginine 342 (R342) was the most frequently mutated amino acid, with 40.0% (6/15) of mutations (p.R342* and p.R342fs*3) occurring in this codon.

Table III. Targeted sequencing results: Tumor protein 53 (TP53) mutations.

IDS: Interval debulking surgery; IHC: immunohistochemical staining; LSO: left salpingo-oophorectomy; NA: not applicable; NAC: Neoadjuvant chemotherapy; NC: negative control; PC: positive control; PDS: primary debulking surgery; POAC: postoperative adjuvant chemotherapy; SDS: secondary debulking surgery.

Five of the nine PDS samples exhibited nonsense mutations (samples 3, 4, 5, 6, and 10-1). Splice site deletions (samples 1, 2, and 8-1) and missense mutations (sample 7) were observed in three and one PDS samples, respectively. All (6/15) the remaining samples obtained by SDS, IDS, or laparoscopic biopsy also showed TP53 mutations. In two patients who underwent neoadjuvant and two who underwent postoperative adjuvant chemotherapy, the TP53 mutations identified in samples obtained after chemotherapy were the same as those detected in pre-chemotherapeutic samples. Two patients had frameshift mutations and the other two had a splice site deletion and a nonsense mutation, respectively. Other mutations classified as variants of uncertain significance are summarized in Table IV (38). The location of the TP53 mutation did not influence tumor location, histological grade, or FIGO stage.

Table IV. Targeted sequencing results: Variants of uncertain significance.

Discussion

Mutation of TP53 is pervasive and characteristic in tubo-ovarian and peritoneal HGSC. p53 has been used as a surrogate marker for the presence of TP53 mutations in HGSC (5). When TP53 is mutated, an aberrant immuno-expression pattern is seen in more than 95% of cases (39). Although a certain number of HGSC cases show wild-type pattern of p53 immunostaining, there have been a few studies showing that, when sequenced, all HGSC cases examined actually bear TP53 mutations (9,15). Considering that p53 immunopositivity is determined by an immunohistochemical staining pattern that correlates with TP53 mutation as opposed to simply a positive or negative staining and that some HGSC cases with truncating or splice site TP53 mutations can show a non-functional p53 expression pattern, the fact that p53 immunostaining cannot accurately predict TP53 mutational status is understandable (40,41).

There are two main chemotherapeutic treatment methods for managing ovarian carcinomas, including HGSC: 1) PDS followed by postoperative platinum-based adjuvant chemotherapy and 2) neoadjuvant chemotherapy followed by IDS. The morphology of post-chemotherapeutic carcinoma is different from its pretreatment appearance, making histological subtyping and grading difficult (40). Meanwhile, despite the tremendous genomic diversity that develops at very early stages of carcinogenesis (42), ancestral clones persist, irrespective of disease progression and chemotherapeutic intervention (43). In this study, we confirmed the presence of TP53 mutations in HGSC showing a wild-type p53 immunostaining pattern and demonstrated the mutational constancy between matched pre- and post-chemotherapeutic HGSC tissue samples. This result is not only concordant with a previous study reporting pre- and post-chemotherapeutic p53 expression in HGSC using only immunohistochemistry (40), but also presents confirmatory evidence via NGS. Our data suggest that similar p53 immunoexpression patterns in pre- and post-chemotherapeutic samples indicate the presence of the same TP53 mutation.

As per our observations, even if an HGSC case presents wild-type p53 immunostaining pattern, there is still a possibility that the tumor bears a TP53 mutation, which could be a non-typical mutation, such as a splice site mutation or frameshift deletion. Furthermore, if an HGSC case bearing a TP53 mutation shows the presence of the mutation before and after chemotherapy, pathologists can make a definite diagnosis of HGSC using NGS targeting TP53 even when the tumor displays wild-type p53 immunostaining pattern or when there are significant morphological alterations induced by chemotherapy. Our results support the notion that TP53 mutations are invariably present in cases of tubo-ovarian and peritoneal HGSC, and that the TP53 mutation persists even after chemotherapy. In HGSC cases with a wild-type p53 immunostaining pattern or with disseminated metastatic lesions of unclear origin after chemotherapy, targeted sequencing using NGS technique can be helpful in revealing the presence of the TP53 mutation.

In conclusion, we observed that all HGSC cases examined exhibited TP53 mutations, confirming the fact that all HGSCs are TP53-mutants even though a subset of HGSCs display a wild-type p53 immunostaining pattern. We did not observe any significant difference in the type of TP53 mutation between paired specimens obtained before and after chemotherapy, suggesting that chemotherapy does not affect TP53 mutational status in tubo-ovarian and peritoneal HGSC.

Conflicts of Interest

None of the Authors have any conflicts of interest to declare regarding this study.

Authors’ Contributions

All Authors were responsible for substantial contributions to the conception and design of the study, acquisition of data, analysis and interpretation of the data, as well as drafting the manuscript, revising the manuscript critically for important intellectual content, and providing final approval of the version to be published.

Acknowledgements

This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (2018R1C1B5043725) and the Basic Science Research Program through the NRF funded by the Ministry of Education (2016R1D1A1B03935584).