Abstract 453: MEK activation is associated with a molecular subgroup in high grade serous ovarian cancer

Introduction

We have previously defined 3 molecular subgroups of High grade serous ovarian cancer (HGSOC), ‘Angio’, ‘Immune’ and ‘Angio_immune’ subgroups using gene expression data from 265 FFPE HGSOC samples obtained from treatment naive patients and who were subsequently treated with platinum-based standard of care (SoC) chemotherapy (carboplatin +/- paclitaxel) (Gourley, et al. J Clin Oncol 32:5s, 2014). Patients within these 3 molecular subgroups respond differently to SOC treatment. The immune subgroup has the best outcome compared to the Angio and Angio_immune subgroups (HR of 0.63 and 0.66 respectively on multivariate analysis). Since it has previously been shown that the MAPK pathway is an important mediator of cisplatin resistance in ovarian cancer, we wanted to investigate if the MAPK pathway was associated with one of the poor prognosis subgroups.

Methods

A gene signature that could detect each of the subgroups Angio, Immune and Angio_immune was generated from the clinical samples. Data from the Cancer Genome Atlas (TCGA) Project was used to test for correlation between each subgroup and phospho-MEK as measured by Reverse Phase Proteomic Array (RPPA). Sensitivity to the MEK inhibitor trametinib (GSK1120212) and cisplatin was determined by 10-day colony formation assay.

Results

We found a statistically significant association between the Angio_immune subgroup signature and Phospho-MEK (serine 217/221) expression (p = 0.047) indicating activation of the MAPK pathway in this subgroup. Additionally we have demonstrated that the Angio_immune subgroup signature is suppressed by MEK inhibition (p = 0.0055) and elevated by KRAS, NRAS and MEK1 overexpression in cell line models (0.0072, 0.0004 and <0.0001). These effects were specific to the Angio_immune subgroup signature as there was no association with the Angio or immune subgroup signatures. Additionally the Angio_immune gene signature could predict resistance to cisplatin and sensitivity to MEK inhibitors in a panel of breast and ovarian cancer cell line models (p = 0.0017 and p = 0.0091). Acquired resistance to cisplatin in cell line models was also associated with MEK activation and an elevated Angio_immune gene signature score, which could be reversed by a MEK inhibitor.

Conclusion

We have identified a molecular subgroup in HGSOC that is associated with MAPK signalling. A gene signature to detect this subgroup from formalin fixed paraffin embedded samples has been developed and predicts sensitivity to MEK inhibitors in pre-clinical model systems. Further work aims to validate the signature in clinical samples from patients treated with a MEK inhibitor.

Citation Format: Nuala McCabe, Charlie Gourley, Andrena McGavigan, Caroline O. Michie, Niamh McGivern, Michael Churchman, Eamonn J. O’Brien, Laura Hill, Timothy S. Davison, Alistair Williams, Glenn McCluggage, Karen E. Keating, Denis P. Harkin, Richard D. Kennedy. MEK activation is associated with a molecular subgroup in high grade serous ovarian cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 453.

Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial

BACKGROUND

The international standard of care for women with suspected advanced ovarian cancer is surgical debulking followed by platinum-based chemotherapy. We aimed to establish whether use of platinum-based primary chemotherapy followed by delayed surgery was an effective and safe alternative treatment regimen.

METHODS

In this phase 3, non-inferiority, randomised, controlled trial (CHORUS) undertaken in 87 hospitals in the UK and New Zealand, we enrolled women with suspected stage III or IV ovarian cancer. We randomly assigned women (1:1) either to undergo primary surgery followed by six cycles of chemotherapy, or to three cycles of primary chemotherapy, then surgery, followed by three more cycles of completion chemotherapy. Each 3-week cycle consisted of carboplatin AUC5 or AUC6 plus paclitaxel 175 mg/m(2), or an alternative carboplatin combination regimen, or carboplatin monotherapy. We did the random assignment by use of a minimisation method with a random element, and stratified participants according to the randomising centre, largest radiological tumour size, clinical stage, and prespecified chemotherapy regimen. Patients and investigators were not masked to group assignment. The primary outcome measure was overall survival. Primary analyses were done in the intention-to-treat population. To establish non-inferiority, the upper bound of a one-sided 90% CI for the hazard ratio (HR) had to be less than 1.18. This trial is registered, number ISRCTN74802813, and is closed to new participants.

FINDINGS

Between March 1, 2004, and Aug 30, 2010, we randomly assigned 552 women to treatment. Of the 550 women who were eligible, 276 were assigned to primary surgery and 274 to primary chemotherapy. All were included in the intention-to-treat analysis; 251 assigned to primary surgery and 253 to primary chemotherapy were included in the per-protocol analysis. As of May 31, 2014, 451 deaths had occurred: 231 in the primary-surgery group versus 220 in the primary-chemotherapy group. Median overall survival was 22.6 months in the primary-surgery group versus 24.1 months in primary chemotherapy. The HR for death was 0.87 in favour of primary chemotherapy, with the upper bound of the one-sided 90% CI 0.98 (95% CI 0.72-1.05). Grade 3 or 4 postoperative adverse events and deaths within 28 days after surgery were more common in the primary-surgery group than in the primary-chemotherapy group (60 [24%] of 252 women vs 30 [14%] of 209, p=0.0007, and 14 women [6%] vs 1 woman [<1%], p=0.001). The most common grade 3 or 4 postoperative adverse event was haemorrhage in both groups (8 women [3%] in the primary-surgery group vs 14 [6%] in the primary-chemotherapy group). 110 (49%) of 225 women receiving primary surgery and 102 (40%) of 253 receiving primary chemotherapy had a grade 3 or 4 chemotherapy related toxic effect (p=0.0654), mostly uncomplicated neutropenia (20% and 16%, respectively). One fatal toxic effect, neutropenic sepsis, occurred in the primary-chemotherapy group.

INTERPRETATION

In women with stage III or IV ovarian cancer, survival with primary chemotherapy is non-inferior to primary surgery. In this study population, giving primary chemotherapy before surgery is an acceptable standard of care for women with advanced ovarian cancer.

FUNDING

Cancer Research UK and the Royal College of Obstetricians and Gynaecologists.

Abstract LB-89: Germ-line and somatic SMARCA4 mutations characterize small cell carcinoma of the ovary, hypercalcemic type

Small-cell carcinoma of the ovary, hypercalcemic type (SCCOHT) is an aggressive tumor and the most common type of undifferentiated ovarian malignancy presenting at less than 40 years of age. Its cause and histogenesis remain unknown. We sequenced the exomes of individuals from three familial cases of SCCOHT. Subsequently, we used whole exome sequencing, Sanger sequencing and immunohistochemistry to analyze germline and tumor DNA from three additional familial cases, 35 non-familial cases and one SCCOHT cell line (BIN-67). DNA sequencing revealed likely deleterious germ-line mutations in the chromatin remodeling gene SMARCA4 in all four familial cases of SCCOHT where DNA was available. This was accompanied by either a somatic mutation or loss of the wild-type allele in the tumor. BIN-67 contained bi-allelic deleterious mutations in SMARCA4. Sequencing of 24 non-familial pathologically-confirmed SCCOHT cases revealed at least one germ-line or somatic likely deleterious SMARCA4 mutation in 22 cases. Immunohistochemical (IHC) analyses of these and an additional 11 tumors showed loss of BRG1 protein in 38/40 cases. Our findings identify alterations in SMARCA4 as a major cause of SCCOHT, which could pave the way for genetic counseling and new treatment approaches.

Citation Format: Leora Witkowski, Jian Carrot-Zhang, Steffen Albrecht, Nancy Hamel, Eva Tomiak, David Grynspan, Emmanouil Saloustros, Catherine Gilpin, Rachel Silva-Smith, François Plourde, Barbara Rivera, Ester Castellsagué, Mona Wu, Somayyeh Fahiminiya, Javad Nadaf, Avi Saskin, Madeleine Arseneault, Rouzan G. Karabakhtsian, Elizabeth A. Reilly, Frederick R. Ueland, Anna Margiolaki, Kitty Pavlakis, Sharon M. Castellino, Janez Lamovec, Lawrence M. Roth, Thomas M. Ulbright, Tracey Bender, Michel Longy, Andrew Berchuck, Marc Tischkowitz, Reiner Siebert, Inga Nagel, Vassilis Georgoulias, Colin J.r. Stewart, Glenn McCluggage, Jocelyne Arseneau, Blaise A. Clarke, Yasser Riazalhosseini, Martin Hasselblatt, Jacek Majewski, William D. Foulkes. Germ-line and somatic SMARCA4 mutations characterize small cell carcinoma of the ovary, hypercalcemic type. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-89. doi:10.1158/1538-7445.AM2014-LB-89

Immunohistochemical detection of p53 and bcl-2 proteins in neoplastic and non-neoplastic endocervical glandular lesions

The study examines p53 and bcl-2 protein expression in a variety of neoplastic and non-neoplastic endocervical glandular lesions. Immunohistochemical staining, using monoclonal antibodies against p53 (DO-7) and bcl-2, was performed on archival paraffin-embedded tissue following microwave antigen retrieval. With DO-7 there was positive nuclear staining in 23/33 cases of adenocarcinoma, 2/10 adenocarcinoma in situ, 2/10 tubo-endometrial metaplasia, 1/10 microglandular hyperplasia, and 1/17 normal endocervix. With adenocarcinoma in situ and non-neoplastic lesions, positive staining was confined to scattered cells. Nine of 23 positive cases of adenocarcinoma showed widespread staining. With anti-bcl-2 antibody, there was positive cytoplasmic staining in 9/33 cases of adenocarcinoma, 0/10 adenocarcinoma in situ, 7/10 tubo-endometrial metaplasia, 1/10 microglandular hyperplasia, and 0/17 normal endocervix. Five of nine positive adenocarcinomas showed widespread staining. There was widespread positive staining for bcl-2 in six of seven positive cases of tubo-endometrial metaplasia. The results indicate that p53 protein expression is frequent in endocervical adenocarcinoma and suggest that mutation of the p53 gene may be important in the evolution of some cases of endocervical adenocarcinoma. Scattered p53-positive cells may be seen in endocervical adenocarcinoma in situ and in non-neoplastic endocervical glandular lesions, the significance of which is uncertain. Bcl-2 protein expression is seen in a proportion of endocervical adenocarcinomas and may play a role in the evolution of these tumors through inhibition of apoptosis. Widespread positivity for bcl-2 protein is seen in most cases of tubo-endometrial metaplasia, suggesting that this type of metaplastic epithelium may represent an unusually stable population of cells.

Basal cell adenocarcinoma of the submandibular gland

Basal cell adenoma is a well recognized histopathologic variant of monomorphic adenoma of salivary gland, but in recent years there have been occasional reports of malignant basal cell tumors of major salivary glands. We present a case report of one such basal cell adenocarcinoma arising in the submandibular gland and discuss the differential diagnosis and distinction from a basal cell adenoma. We also review the published literature on basal cell adenocarcinoma. This is the first case of basal cell adenocarcinoma to be studied in detail by both immunohistochemistry and electron microscopy. There has been debate in the past as to the cell of origin of salivary gland basal cell tumors and specifically as to whether myoepithelial differentiation occurs in these tumors. The immunohistochemical and ultrastructural features of this case provide evidence for a dual population of ductal and myoepithelial cells. A flow cytometric analysis of nuclear deoxyribonucleic acid content showed the presence of DNA aneuploidy within one tissue block.

Grading cervical dysplasia with AgNORs using a semiautomated image analysis system

Colposcopic biopsies were classified according to previously established criteria by a group of three pathologists interested in cervical pathology. Ten cases were identified in each of the following five groups: normal, koilocytosis, low grade squamous intraepithelial lesions (CIN 1), high grade squamous intraepithelial lesions (CIN 2) and high grade squamous intraepithelial lesions (CIN 3). The Crocker technique was used to stain the sections cut 3 microns thick. With ths silver stain the nucleolar organizer regions (NORs) are stained black and referred to as AgNORs. It has been shown that malignant and premalignant changes in cells produce an increase in AgNORs. In each case eight images were captured using a 100x oil-immersion objective and stored in a Datacube Maxvideo system as 512 x 480 pixels in an 8-bit grayscale per image. The images were processed using the NeoPath field-of-view computer to detect the AgNORs and nuclei by using grayscale mathematical morphology algorithms. Color overlays of the AgNORs and nuclei were created using segmentation algorithms. The results show that it is possible to differentiate between low grade squamous intraepithelial lesions (CIN 1) and high grade squamous intraepithelial lesions (CIN 2 and CIN 3) taken together; however, there is no difference between low grade squamous intraepithelial lesions (CIN 1) and koilocytosis. The results support the concept that dysplasia cannot be classified effectively into three grades and that low grade squamous intraepithelial lesions (mild dysplasia [CIN 1]) is indistinguishable from koilocytosis.

Value of the necropsy in perioperative deaths

A series of 213 perioperative deaths was studied out of a total of 1451 consecutive necropsies carried out over three years. Discrepancies between the clinical and the necropsy diagnosis were assessed under four classes of discrepant diagnosis: class I, patient survival affected, treatable; class II, patient survival affected but not treatable; class III, correlated to cause of death but treatable; and class IV, incidental diagnosis which could not have been made before death. Major discrepancies of classes I and II were found in 44 (21%) and 62 (29%) cases, respectively. Minor discrepancies of classes III and IV were found in 63 (30%) and 101 (47%) cases, respectively. No discrepancies were found in 50 (23.5%) cases. These results confirm the continuing value of the necropsy in the assessment of perioperative deaths.