Abstract
Objective:
We sought to determine the predictive value of combining tumor molecular subtype and computerized tomography (CT) imaging for surgical outcomes after primary cytoreductive surgery in advanced stage high-grade serous ovarian cancer (HGSOC) patients.
Methods:
We identified 129 HGSOC patients who underwent pre-operative CT imaging and post-operative tumor mRNA profiling. A continuous CT-score indicative of overall disease burden was defined based on six imaging measurements of anatomic involvement. Molecular subtypes were derived from mRNA profiling of chemo-naïve tumors and classified as mesenchymal (MES) subtype (36%) or non-MES subtype (64%). Fischer exact tests and multivariate logistic regression examined residual disease and surgical complexity.
Results:
Women with higher CT-scores were more likely to have MES subtype tumors (p = 0.014). MES subtypes and a high CT-score were independently predictive of macroscopic disease and high surgical complexity. In multivariate models adjusting for age, stage and American Society of Anesthesiologists (ASA) score, patients with a MES subtype and high CT-score had significantly elevated risk of macroscopic disease (OR = 26.7, 95% CI = [6.42, 187]) and were more likely to undergo high complexity surgery (OR = 9.53, 95% CI = [2.76, 40.6], compared to patients with non-MES tumor and low CT-score.
Conclusion:
Preoperative CT imaging combined with tumor molecular subtyping can identify a subset of women unlikely to have resectable disease and likely to require high complexity surgery. Along with other clinical factors, these may refine predictive scores for resection and assist treatment planning. Investigating methods for pre-surgical molecular subtyping is an important next step.
Introduction
The most common ovarian cancer histotype, high-grade serous ovarian cancer (HGSOC), is usually diagnosed at an advanced stage, and primary surgery is recommended when a complete resection can safely be performed (1-3). Multiple studies have confirmed that residual disease after primary surgery is highly predictive of survival, and no gross macroscopic disease is associated with the best oncologic outcomes. An alternative to upfront surgery is neoadjuvant chemotherapy (NACT), commonly due to an inability to achieve complete or near complete resection. To avoid futile surgical attempts, and, importantly, to avoid withholding surgery that may be successful, accurate predictive models of resectability and anticipated surgical complexity are needed. Currently available models such as computerized tomography (CT) imaging or laparoscopic evaluation, while useful, still have room for improvement, and their predictive potential depends on center experience and the use of complex surgical procedures (4-7). Laparoscopic scoring procedures, while less invasive than laparotomy are associated with discomfort, surgical risks and significant costs. There remains a clinical need to develop safer and better predictors for resectability in advanced HGSOC and divert these patients to alternative initial treatment strategies.
Based on tumor transcriptome profiling, The Cancer Genome Atlas (TCGA) Research Network and others has reported that HGSOC can be generally classified into molecular subtypes. The most distinctive subtype is mesenchymal (MES) which is characterized by high expression of extracellular matrix components, desmoplastic stroma, and poor clinical outcomes (8-9). We previously reported that women with MES tumors were more likely to have upper abdominal disease and higher rates of gross residual disease, despite requiring higher complexity surgery for cytoreduction (10)(11). Further, MES subtype tumors are enriched in the upper abdominal disease sites (12).
Hypothesizing that molecular subtype would be additive to imaging for the prediction of disease resectability, we conducted an observational pilot study. We sought to determine if CT imaging combined with tumor molecular subtype predicts successful (e.g. complete gross resection) cytoreductive surgery in patients with stage IIIC/IV HGSOC.
Patients and Methods
Patient population
A cohort of 129 stage IIIC/IV HGSOC patients treated with primary cytoreductive surgery followed by platinum-based adjuvant chemotherapy at the Mayo Clinic (Rochester, MN) between 2003 and 2011 were studied. Inclusion criteria were patients had curative-intent primary surgery, an available pre-operative CT scan, and tumor tissues for molecular subtyping. Clinical data were collected retrospectively from the medical record. Residual disease after surgery was categorized as no macroscopic residual disease (defined as RD0 for this study), gross disease ≤1 cm (RD1), and gross disease >1 cm (RD2). Surgical complexity was defined as low/intermediate or high as described previously (13). All participants provided written informed consent, and the study was approved by the Mayo Clinic institutional review boards.
CT evaluation criteria and CT-score
Pre-operative CT scans (within 90 days before surgery) were analyzed for anatomic distribution of disease (4). Based on prior studies six anatomic features were considered in a binary fashion as present (score=1) or absent (score=0): (a) involvement of diaphragm disease, (b) existence of gastrohepatic/porta hepatis lesions, (c) root of superior mesenteric artery, (d) presence of moderate to severe amount of ascites, (e) involvement of intrahepatic lesion, and (f) diffuse peritoneal thickening (7). Scores for the 6 radiographic factors were summed to create a continuous CT-score of disease burden ranging from 0 (lowest) to 6 (highest) which were analyzed in continuous and dichotomized forms (CT-low: 0 or 1; CT-high: 2 to 6).
Tumor expression-based molecular subtype characterization
Following surgical resection, tissues were reviewed by an experienced gynecologic pathologist (GLK), who confirmed the presence of at least 70% tumor content; 104 tumors were sampled from known sites including 80 tumors from primary cancer sites (ovary, fallopian tube) and 24 from metastatic sites. For the majority of cases, fresh frozen tumors were subjected to Agilent Whole Human Genome 4 × 44 K Expression Arrays (N=51) or RNA-sequencing (N=22); other cases used archival tissue and NanoString nCounter technology (N=56) (14-19). Batch-effects among RNA preparation and profiling experiments were corrected for as previously described (14-16), and molecular subtype assignments were derived separately per previously published studies on each separate RNA profiling technology (11, 19-20). For cases on whom data from multiple transcriptome platforms were available, we considered the descending expression measurement reliability of two-channel expression profiling from fresh-frozen tumors (Agilent), sequencing measures from fresh-frozen tumors (RNA-sequencing), to customized assays quantifying expression levels from clinical archival tissues (Nanostring); thus, we prioritized use of molecular subtype assignments based on Agilent microarray, then RNA-sequencing, then NanoString data.
Statistical analyses
Evaluation of pairwise associations between the CT-score of disease burden (CT-low, CT-high), molecular subtype (MES, non-MES), tumor collection site (primary, metastasis; N=104), and surgical outcomes (RD0, RD1, RD2) used Fisher’s exact testing. In a subset of 118 optimally debulked cases (i.e., RD0 and RD1 combined), surgical complexity (low/intermediate versus high) was also analyzed in a pairwise manner. Multivariable logistic regression models were fit to estimate MES subtype (MES versus non-MES) odds-ratio (ORs) and corresponding 95% confidence intervals (CIs) for predicting an RD0 surgical result (versus RD1/RD2); covariates were those with known or suspected associations with surgical outcome: age (<70, >=70), American Society of Anesthesiologists (ASA) class (2, 3), FIGO stage (IIIC, IV), and CT-score (continuous and binary). In addition, CT-score and MES were combined into a predictive variable, and sensitivity analyses limited to adnexal-sourced tissues. Statistical analysis was performed using the R software. All calculated P values were two-sided, and P values <0.05 were considered statistically significant.
Results
Residual Disease
Among 129 patients with stage IIIC-IV HGSOC, 36% of patients were found to have MES tumors and 64% had non-MES subtype tumors, consistent with expected ratios (Table 1). MES molecular subtype (MES versus non-MES) and continuous CT-score were significantly associated with each other(Fisher’s exact test p = 0.014), such that patients with MES tumors showed greater CT-scores of disease burden. According to multivariate logistic regression analysis adjusting for age (<70, >=70), ASA-score, and stage (IV versus III), dichotomized CT-score, and MES subtype were independently predictive of leaving macroscopic residual disease (RD1 or RD2) after primary surgery: CT-high v CT-low p<0.001 (OR = 6.71, 95% CI = [2.9, 16.5]), MES v non-MES p=0.016 (OR = 3, 95% CI = [1.25, 7.58]) (Fig. 1). The results were consistent after sensitivity analyses of 104 cases after adjusting for site of sampling (ovary/tube versus metastasis).
Table 1.
Patient and HGSOC tumor characteristics (N=129)
| All patients (N=129) |
Mesenchymal (N=46) |
Non-Mesenchymal N=83 |
|
|---|---|---|---|
| Age (years), mean (SD) | 64.6 (10.7) | 66 (11.5) | 64 (10.7) |
| FIGO stage | |||
| IIIC | 90 (70%) | 29 (63%) | 61 (73%) |
| IV | 39 (30%) | 17 (37%) | 22 (27%) |
| ASA class | |||
| 2 | 63 (49%) | 18 (39%) | 37 (54%) |
| 3 | 66 (51%) | 28 (61%) | 38 (46%) |
| Tumor molecular subtype collection site | |||
| Adnexa | 80 (60%) | 23 (50%) | 57 (69%) |
| Omentum or other metastatic site | 14 (8%) | 13 (28%) | 11 (13%) |
| Unknown | 25 (19%) | 10 (22%) | 15 (18%) |
| CT-score of disease burden | |||
| Ordinal, mean (SD) | 1.6 (1.4) | 1.9 (1.3) | 1.4 (1.4) |
| CT-low (CT score = 0 or 1) | 70 (54%) | 21 (46%) | 49 (59%) |
| CT-high (CT score =2 or above) | 59 (46%) | 25 (54%) | 34 (41%) |
| Surgical complexity | |||
| Low/Intermediate | 64 (50%) | 17 (37%) | 47 (57%) |
| High | 65 (50%) | 29 (63%) | 36 (43%) |
| Surgical outcome | |||
| No macroscopic disease (RD0) | 59 (46%) | 13 (28%) | 46 (55%) |
| Residual disease =< 1cm (RD1) | 59 (46%) | 27 (59%) | 30 (36%) |
| Residual disease > 1cm (RD2) | 11 (8%) | 6 (13%) | 7 (8%) |
Abbreviations: ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics. The combined group was jointly determined according to CT-based disease burden (if CT-score<2: CT-low, otherwise CT-high) and tumor’s molecular subtype (C1.MES subtype or non-C1.MES subtype).
Figure 1:
Forest plot based on multivariate analysis predicting macroscopic residual disease (RD1 or RD2) at primary cytoreductive surgery. All factors shown are included in model. Mesenchymal subtype and CT score high were independently predictive of having residual disease after surgery. N=129 HGSOC.
We created four subgroups based on CT-score and MES molecular subtype: CT-low/non-MES (n=49; 38%), CT-low/MES (n=21; 16%), CT-high/non-MES (n=34; 26%) and CT-high/MES (n=25; 19%). CT-molecular group was, as expected, significantly associated with residual disease status (RD0, RD1 or RD2), with the CT-high + MES group being completely resected in only 8% of cases, contrasting with complete resection in 71.4% of CT-low + non-MES patients (Fig. 2). Multivariate logistic regression confirmed that patients in the CT-high + MES group had an elevated risk of macroscopic disease (OR = 26.73, 95% CI = [6.42, 186.94], p<0.001), comparing to patients in the CT-low and non-MES group as reference, adjusted for age (<70, >=70), stage (IV versus III), and ASA-score (Fig 3).
Figure 2:
Among all the 129 patients, CT-Molecular subtype had statistically significant association with residual disease status (p<0.001), with CT-high + MES group had the lowest RD0 proportion.
Figure 3:
According to logistic regression predictive of non-RD0 status, CT-molecular subtype had independent predictive values after adjusting for age, ASA score (=3) and stage (IIIC versus IV). N=129 HGSOC.
Surgical complexity
When considering only the subset of optimally debulked patients (N=118, RD0 or RD1; <=1cm), CT-molecular patient groups showed statistically significant association with surgical complexity (high versus low/moderate, p<0.001); patients in the CT-high and MES cohort had the highest proportion of high-complexity surgeries compared to the patients in the CT-low and non-MES group (81% versus 35%). Adjusting for age, ASA and stage, patients in the CT-high + MES group had a markedly elevated risk of undergoing high complexity surgery (OR = 9.53, 95% CI = [2.76, 40.63], p<0.001), when comparing to those in the CT-low and non- MES group as reference, despite the low rate of complete resection in these high-risk cases (Fig. 4). Altogether, CT and molecular subtype independently predict surgical complexity among optimally debulked cases (data not shown).
Figure 4:
Among 118 patients with optimal primary cytoreduction, forest plot predicting need for high surgical complexity, adjusted for age > 70 years old, stage IIIC versus IV, and ASA.
Evaluation by sampling site
Given differences in molecular subtype depending on the sampling site, we examined the different frequency of subtype in different sampling sites and effects on surgical outcomes. As a sensitivity analysis assessing impacts of sampling source used in molecular subtyping, a series of stratified analyses were done using 80 patients with primary tumors (ovary, fallopian tube) as the tissue source for molecular subtype assignments; all of these patients were optimally debulked. MES subtype trended towards association with CT scores, although it was statistically insignificant (p=0.14). Among these adnexal primary-sampled patients, CT and molecular subtype were again independently associated with RD0, and, as expected, multivariate logistic regression suggested that CT-high + MES group remained the highest risk of macroscopic disease (OR = 32.32, 95% CI = [5.11, 644.74], p=0.002), shown as Supplementary Fig. 1. Similarly, the CT-high + MES group were more likely to undergo high complexity surgery, (OR = 13.9, 95%CI = [2.05-286.68], p=0.022), Supplementary Fig. 2. These trends were also apparent among 24 cases sampled at metastases (data not shown).
Discussion
We report that the combination of MES molecular subtype and higher disease burden based on defined CT characteristics is associated with very low likelihood of complete gross resection of ovarian cancer in an aggressive surgical center. This result is independent of age, ASA score, and stage, but investigating how all of these clinical and pathologic characteristics interact will be important. Sample site of tumor used for molecular subtyping did not appear to alter the results. We previously reported that MES subtype is a poor prognosis factor and correlates with higher residual disease after primary cytoreductive surgery (11). The present study builds on this research and suggests a strategy to improve preoperative prediction of surgical outcome by including imaging and molecular subtype in predictive models, especially if we can successfully define subtype based on pre-operative biopsy. This is an important next step of our prospective work in our institution. We observe that patients with MES subtype tumors with a high CT score, have a significantly lower chance of RD0 (8%) despite having the highest percentage of high-complexity surgeries (81%). Currently preoperative imaging is available, and if pre-operative molecular subtyping could be successfully obtained through diagnostic biopsy, this approach could improve our treatment of primary ovarian cancer by avoiding procedures with the lowest likelihood of success.
Molecular characterizations of ovarian cancer have been studied extensively for HGSOC (9, 11). MES subtype tumors are reproducible across different studies and uniquely characterized by up-regulations of stromal- and fibroblast-specific genes, such as FAP, FABP4, POSTN, COL11A1, and VIM. MES subtype has been reported to have activations of TGF-beta and AXL-GAS6 pathways, and some of these pathway findings may lead to subtype-specific treatment options (21). MES subtype was reportedly associated with higher percentage of stromal content (22), distinct desmoplastic reactions in pathological reviews (23), and higher amounts of intraperitoneal disease spread (10), all supporting it as an unique group of HGSOC tumors with distinct molecular biology.
To our knowledge, this study represents the first study to integrate molecular subtype and CT data to predict surgical outcomes, albeit a small sample size. Although CT-imaging are routinely used in clinic for HGSOC diagnosis and disease evaluation, only a few studies examine the relationships between CT and molecular data, for example: Vargas et al. examined CT features’ relationships with TCGA subtypes in 46 ovarian cancer cases and found MES subtype was associated with mesenteric infiltration (24); the same research group also later conducted an expanded study in 92 HGSOC patients to predict time-to-disease progression (25). Similarly, a Japanese research group attempted to infer molecular subtypes using CT imaging features in 65 ovarian cancer cases with limited sensitivity (67%) (26). These prior CT imaging and molecular associations studies and our reported study suggest that prospective evaluation of molecular subtype from diagnostic biopsies and pre-operative imaging would be a valuable next step. A strength of our report is the ability to disern RD0 disease from RD1 and RD2. An important area of subsequent research will be conducting molecular subtyping on pre-operative biopsies or cytology collection.
Our current study has the limitation that it was done retrospectively in a single center; it will be desirable to examine the proposed CT-molecular grouping in other surgical centers, and to validate associations with surgical outcome independently. As another limitation, the molecular subtypes were inferred from expression profiling measured in surgically resected tumors, which may not be available for real clinical scenarios before surgeries. Future work can expand to larger collections of archival tissues (12, 15, 19), which are tissue sources clinically feasible for preoperative assessments. Molecular subtyping was performed across multiple platforms, with the Agilent microarray being the most commonly used due to the availablility of data. More work is needed to understand the consistency of subtyping between different RNA platforms, and to consider tissue source and block differences.
In summary, molecular subtype and estimates of CT-based disease burden appear to be complementary markers for surgical resectability. Future studies are warranted to address remaining obstacles including i) rapid molecular classification from preoperative biopsy, ii) most valid site(s) for sampling. This trajectory of research can facilitate continued individualization of care based on both biology and surgical resectability.
Supplementary Material
Supplementary Figure 1: Among 80 patients with adnexal sampling site, CT-molecular subtype had independent predictive values after adjusting for age, ASA score (=3), stage (=IV).
Supplementary Figure 2: Multivariate prediction of high surgical complexity among optimal debulking patients with primary tumors as molecular subtype tissue sources (n=80)
HIGHLIGHTS.
Cases with high disease burden defined by CT score were more likely to have mesenchymal subtype tumors
Mesenchymal subtype and high CT score were associated with the lowest rate of complete resection
Among optimally debulked patients, those with mesenchymal subtype and high CT-score had the highest surgical complexity
Funding sources:
National Institutes of Health/National Cancer Institute (NCI) Grants to SJR [grant number R01CA172404
Canadian Institutes for Health Research (Proof-of-Principle I program, no grant number applicable)
United States Department of Defense Ovarian Cancer Research Program [grant number OC110433].
SJR is supported by National Health and Medical Research Council of Australia (NHMRC) grant APP2009840.
The contents of the published material are solely the responsibility of the authors and do not reflect the views of NHMRC.
Footnotes
Conflicts of Interests
The remaining authors have no conflicts of interest to declare.
References
- 1.Reid BM, Permuth JB and Sellers TA (2017). "Epidemiology of ovarian cancer: a review." Cancer Biol Med 14(1): 9–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wright AA, Bohlke K, Armstrong DK, Bookman MA, Cliby WA, Coleman RL, Dizon DS, Kash JJ, Meyer LA, Moore KN, Olawaiye AB, Oldham J, Salani R, Sparacio D, Tew WP, Vergote I, Edelson MI, Neoadjuvant Chemotherapy for Newly Diagnosed, Advanced Ovarian Cancer: Society of Gynecologic Oncology and American Society of Clinical Oncology Clinical Practice Guideline, J Clin Oncol. 34 (2016) 3460 3473. 10.1200/jco.2016.68.6907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Himoto Y, Cybulska P, Shitano F, Sala E, Zheng J, Capanu M, Nougaret S, Nikolovski I, Vargas HA, Wang W, Mueller JJ, Chi DS and Lakhman Y (2019). "Does the method of primary treatment affect the pattern of first recurrence in high-grade serous ovarian cancer?" Gynecol Oncol 155(2): 192–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Suidan RS, Ramirez PT, Sarasohn DM, Teitcher JB, Iyer RB, Zhou Q, Iasonos A, Denesopolis J, Zivanovic O, Roche KCL, Sonoda Y, Coleman RL, Abu-Rustum NR, Hricak H, Chi DS, A multicenter assessment of the ability of preoperative computed tomography scan and Ca-125 to predict gross residual disease at primary debulking for advanced epithelial ovarian cancer, Gynecol Oncol. 145 (2017) 27 31. 10.1016/j.ygyno.2017.02.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fagotti A, Vizzielli G, FANFANI F, Costantini B, Ferrandina G, Gallotta V, Alletti SG, Tortorella L, SCAMBIA G, Introduction of staging laparoscopy in the management of advanced epithelial ovarian, tubal and peritoneal cancer: Impact on prognosis in a single institution experience, Gynecol Oncol. 131 (2013) 341 346. 10.1016/j.ygyno.2013.08.005. [DOI] [PubMed] [Google Scholar]
- 6.Axtell AE, Lee MH, Bristow RE, Dowdy SC, Cliby WA, Raman S, Weaver JP, Gabbay M, Ngo M, Lentz S, Cass I, Li AJ, Karlan BY, Holschneider CH, Multi-Institutional Reciprocal Validation Study of Computed Tomography Predictors of Suboptimal Primary Cytoreduction in Patients With Advanced Ovarian Cancer, J Clin Oncol. 25 (2007) 384–389. 10.1200/jco.2006.07.7800. [DOI] [PubMed] [Google Scholar]
- 7.Kumar A, Sheedy S, Kim B, Suidan R, Sarasohn DM, Nikolovski I, Lakhman Y, McGree ME, Weaver AL, Chi D, Cliby WA, Models to predict outcomes after primary debulking surgery: Independent validation of models to predict suboptimal cytoreduction and gross residual disease, Gynecol Oncol. (2019). 10.1016/j.ygyno.2019.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tothill RW, Tinker AV, George J, Brown R, Fox SB, Lade S, Johnson DS, Trivett MK, Etemadmoghadam D, Locandro B, Traficante N, Fereday S, Hung JA, Chiew Y-E, Haviv I, A.O.C.S. Group, Gertig D, deFazio A, Bowtell DDL, Novel Molecular Subtypes of Serous and Endometrioid Ovarian Cancer Linked to Clinical Outcome, Clin Cancer Res. 14 (2008) 5198–5208. 10.1158/1078-0432.ccr-08-0196. [DOI] [PubMed] [Google Scholar]
- 9.Verhaak RGW, Tamayo P, Yang J-Y, Hubbard D, Zhang H, Creighton CJ, Fereday S, Lawrence M, Carter SL, Mermel CH, Kostic AD, Etemadmoghadam D, Saksena G, Cibulskis K, Duraisamy S, Levanon K, Sougnez C, Tsherniak A, Gomez S, Onofrio R, Gabriel S, Chin L, Zhang N, Spellman PT, Zhang Y, Akbani R, Hoadley KA, Kahn A, Köbel M, Huntsman D, Soslow RA, Defazio A, Birrer MJ, Gray JW, Weinstein JN, Bowtell DD, Drapkin R, Mesirov JP, Getz G, Levine DA, Meyerson M, T.C.G.A.R. Network, Prognostically relevant gene signatures of high-grade serous ovarian carcinoma, J Clin Invest. 123 (2013) 517–525. 10.1172/jci65833. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Torres D, Kumar A, Wallace SK, Bakkum-Gamez JN, Konecny GE, Weaver AL, McGree ME, Goode EL, Cliby WA and Wang C (2017). "Intraperitoneal disease dissemination patterns are associated with residual disease, extent of surgery, and molecular subtypes in advanced ovarian cancer." Gynecologic oncology 147(3): 503–508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wang C, Armasu SM, Kalli KR, Maurer MJ, Heinzen EP, Keeney GL, Cliby WA, Oberg AL, Kaufmann SH and Goode EL (2017). "Pooled Clustering of High-Grade Serous Ovarian Cancer Gene Expression Leads to Novel Consensus Subtypes Associated with Survival and Surgical Outcomes." Clinical cancer research. 23(15): 4077–4085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Talhouk A, Kommoss S, Mackenzie R, Cheung M, Leung S, Chiu DS, Kalloger SE, Huntsman DG, Chen S, Intermaggio M, Gronwald J, Chan FC, Ramus SJ, Steidl C, Scott DW and Anglesio MS (2016). "Single-Patient Molecular Testing with NanoString nCounter Data Using a Reference-Based Strategy for Batch Effect Correction." PloS one 11(4): e0153844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Aletti GD, Dowdy SC, Podratz KC and Cliby WA (2007). "Relationship among surgical complexity, short-term morbidity, and overall survival in primary surgery for advanced ovarian cancer." American journal of obstetrics and gynecology 197(6): 676 e671–677. [DOI] [PubMed] [Google Scholar]
- 14.Goode EL, DeRycke M, Kalli KR, Oberg AL, Cunningham JM, Maurer MJ, Fridley BL, Armasu SM, Serie DJ, Ramar P, Goergen K, Vierkant RA, Rider DN, Sicotte H, Wang C, Winterhoff B, Phelan CM, Schildkraut JM, Weber RP, Iversen E, Berchuck A, Sutphen R, Birrer MJ, Hampras S, Preus L, Gayther SA, Ramus SJ, Wentzensen N, Yang HP, Garcia-Closas M, Song H, Tyrer J, Pharoah PP, Konecny G, Sellers TA, Ness RB, Sucheston LE, Odunsi K, Hartmann LC, Moysich KB and Knutson KL (2013). "Inherited variants in regulatory T cell genes and outcome of ovarian cancer." PloS one 8(1): e53903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Konecny GE, Wang C, Hamidi H, Winterhoff B, Kalli KR, Dering J, Ginther C, Chen HW, Dowdy S, Cliby W, Gostout B, Podratz KC, Keeney G, Wang HJ, Hartmann LC, Slamon DJ and Goode EL (2014). "Prognostic and therapeutic relevance of molecular subtypes in high-grade serous ovarian cancer." Journal of the National Cancer Institute 106(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang C, Winterhoff BJ, Kalli KR, Block MS, Armasu SM, Larson MC, Chen HW, Keeney GL, Hartmann LC, Shridhar V, Konecny GE, Goode EL and Fridley BL (2016). "Expression signature distinguishing two tumour transcriptome classes associated with progression-free survival among rare histological types of epithelial ovarian cancer." British journal of cancer 114(12): 1412–1420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Earp MA, Raghavan R, Li Q, Dai J, Winham SJ, Cunningham JM, Natanzon Y, Kalli KR, Hou X, Weroha SJ, Haluska P, Lawrenson K, Gayther SA, Wang C, Goode EL and Fridley BL (2017). "Characterization of fusion genes in common and rare epithelial ovarian cancer histologic subtypes." Oncotarget 8(29): 46891–46899. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Millstein J, Budden T, Goode EL, Anglesio MS, Talhouk A, Intermaggio MP, Leong HS, Chen S, Elatre W, Gilks B, Nazeran T, Volchek M, Bentley RC, Wang C, Chiu DS, Kommoss S, Leung SCY, Senz J, Lum A, Chow V, Sudderuddin H, Mackenzie R, George J, A. Group, Bowtell D, Chenevix-Trench G, Green A, Webb P, DeFazio A, Gertig D, Traficante N, Fereday S, Moore S, Hung J, Harrap K, Sadkowsky T, Pandeya N, Malt M, Mellon A, Robertson R, Bergh TV, Jones M, Mackenzie P, Maidens J, Nattress K, Chiew YE, Stenlake A, Sullivan H, Alexander B, Ashover P, Brown S, Corrish T, Green L, Jackman L, Ferguson K, Martin K, Martyn A, Ranieri B, White J, Jayde V, Mamers P, Bowes L, Galletta L, Giles D, Hendley J, Alsop K, Schmidt T, Shirley H, Ball C, Young C, Viduka S, Tran H, Bilic S, Glavinas L, Brooks J, Stuart-Harris R, Kirsten F, Rutovitz J, Clingan P, Glasgow A, Proietto A, Braye S, Otton G, Shannon J, Bonaventura T, Stewart J, Begbie S, Friedlander M, Bell D, Baron-Hay S, Ferrier,a A, Gard G, Nevell D, Pavlakis N, Valmadre S, Young B, Camaris C, Crouch R, Edwards L, Hacker N, Marsden D, Robertson G, Beale P, Beith J, Carter J, Dalrymple C, Houghton R, Russell P, Links M, Grygiel J, Hill J, Brand A, Byth K, Jaworski R, Harnett P, Sharma R, Wain G, Ward B, Papadimos D, Crandon A, Cummings M, Horwood K, Obermair A, Perrin L, Wyld D, Nicklin J, Davy M, Oehler MK, Hall C, Dodd T, Healy T, Pittman K, Henderson D, Miller J, Pierdes J, Blomfield P, Challis D, McIntosh R, Parker A, Brown B, Rome R, Allen D, Grant P, Hyde S, Laurie R, Robbie M, Healy D, Jobling T, Manolitsas T, McNealage J, Rogers P, Susil B, Sumithran E, Simpson I, Phillips K, Rischin D, Fox S, Johnson D, Lade S, Loughrey M, O’Callaghan N, Murray W, Waring P, Billson V, Pyman J, Neesham D, Quinn M, Underhill C, Bell R, Ng LF, Blum R, Ganju V, Hammond I, Leung Y, McCartney A, Buck M, Haviv I, Purdie D, Whiteman D, Zeps N, Fereday S, Hendley J, Traficante N, Steed H, Koziak JM, Köbel M, McNeish IA, Goranova T, Ennis D, Macintyre G, Silva DSD, y Cajal TR, García-Donas J, Polo SH, Rodriguez GC, Cushing-Haugen KL, Harris HR, Greene CS, Zelaya RA, Behrens S, Fortner RT, Sinn P, Herpel E, Lester J, Lubiński J, Oszurek O, Tołoczko A, Cybulski C, Menkiszak J, Pearce CL, Pike MC, Tseng C, Alsop J, Rhenius V, Song H, Jimenez-Linan M, Piskorz AM, Gentry-Maharaj A, Karpinskyj C, Widschwendter M, Singh N, Kennedy CJ, Sharma R, Harnett PR, Gao B, Johnatty SE, Sayer R, Boros J, Winham SJ, Keeney GL, Kaufmann SH, Larson MC, Luk H, Hernandez BY, Thompson PJ, Wilkens LR, Carney ME, Trabert B, Lissowska J, Brinton L, Sherman ME, Bodelon C, Hinsley S, Lewsley LA, Glasspool R, Banerjee SN, Stronach EA, Haluska P, Ray-Coquard I, Mahner S, Winterhoff B, Slamon D, Levine DA, Kelemen LE, Benitez J, Chang-Claude J, Gronwald J, Wu AH, Menon U, Goodman MT, Schildkraut JM, Wentzensen N, Brown R, Berchuck A, Chenevix-Trench G, deFazio A, Gayther SA, García MJ, Henderson MJ, Rossing MA, Beeghly-Fadiel A, Fasching PA, Orsulic S, Karlan BY, Konecny GE, Huntsman DG, Bowtell DD, Brenton JD, Doherty JA, Pharoah PDP, Ramus SJ, Prognostic gene expression signature for high-grade serous ovarian cancer, Ann Oncol Official J European Soc Medical Oncol. 31 (2020) 1240–1250. 10.1016/j.annonc.2020.05.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Talhouk A, George J, Wang C, Budden T, Tan TZ, Chiu DS, Kommoss S, Leong HS, Chen S, Intermaggio MP, Gilks B, Nazeran TM, Volchek M, Elatre W, Bentley RC, Senz J, Lum A, Chow V, Sudderuddin H, Mackenzie R, Leong SCY, Liu G, Johnson D, Chen B, A. Group, Alsop J, Banerjee SN, Behrens S, Bodelon C, Brand AH, Brinton L, Carney ME, Chiew Y-E, Cushing-Haugen KL, Cybulski C, Ennis D, Fereday S, Fortner RT, García-Donas J, Gentry-Maharaj A, Glasspool R, Goranova T, Greene CS, Haluska P, Harris HR, Hendley J, Hernandez BY, Herpel E, Jimenez-Linan M, Karpinskyj C, Kaufmann SH, Keeney GL, Kennedy CJ, Köbel M, Koziak JM, Larson MC, Lester J, Lewsley L-A, Lissowska J, Lubiński J, Luk H, Macintyre G, Mahner S, McNeish IA, Menkiszak J, Nevins N, Osorio A, Oszurek O, Palacios J, Hinsley S, Pearce CL, Pike MC, Piskorz AM, Ray-Coquard I, Rhenius V, Rodriguez-Antona C, Sharma R, Sherman ME, Silva DD, Singh N, Sinn P, Slamon D, Song H, Steed H, Stronach EA, Thompson PJ, Tołoczko A, Trabert B, Traficante N, Tseng C-C, Widschwendter M, Wilkens LR, Winham SJ, Winterhoff B, Beeghly-Fadiel A, Benitez J, Berchuck A, Brenton JD, Brown R, Chang-Claude J, Chenevix-Trench G, deFazio A, Fasching PA, García MJ, Gayther SA, Goodman MT, Gronwald J, Henderson MJ, Karlan BY, Kelemen LE, Menon U, Orsulic S, Pharoah PDP, Wentzensen N, Wu AH, Schildkraut JM, Rossing MA, Konecny GE, Huntsman DG, Huang RY-J, Goode EL, Ramus SJ, Doherty JA, Bowtell DD, Anglesio MS, Development and Validation of the Gene Expression Predictor of High-grade Serous Ovarian Carcinoma Molecular SubTYPE (PrOTYPE), Clin Cancer Res. 26 (2020) 5411–5423. 10.1158/1078-0432.ccr-20-0103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Konecny GE, Wang C, Hamidi H, Winterhoff B, Kalli KR, Dering J, Ginther C, Chen H-W, Dowdy S, Cliby W, Gostout B, Podratz KC, Keeney G, Wang H-J, Hartmann LC, Slamon DJ, Goode EL, Prognostic and Therapeutic Relevance of Molecular Subtypes in High-Grade Serous Ovarian Cancer, J National Cancer Inst. 106 (2014). 10.1093/jnci/dju249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Antony J, Tan TZ, Kelly Z, Low J, Choolani M, Recchi C, Gabra H, Thiery JP and Huang RY (2016). "The GAS6-AXL signaling network is a mesenchymal (Mes) molecular subtype-specific therapeutic target for ovarian cancer." Science signaling 9(448): ra97. [DOI] [PubMed] [Google Scholar]
- 22.Verhaak RG, Tamayo P, Yang JY, Hubbard D, Zhang H, Creighton CJ, Fereday S, Lawrence M, Carter SL, Mermel CH, Kostic AD, Etemadmoghadam D, Saksena G, Cibulskis K, Duraisamy S, Levanon K, Sougnez C, Tsherniak A, Gomez S, Onofrio R, Gabriel S, Chin L, Zhang N, Spellman PT, Zhang Y, Akbani R, Hoadley KA, Kahn A, Kobel M, Huntsman D, Soslow RA, Defazio A, Birrer MJ, Gray JW, Weinstein JN, Bowtell DD, Drapkin R, Mesirov JP, Getz G, Levine DA and Meyerson M (2013). "Prognostically relevant gene signatures of high-grade serous ovarian carcinoma." The Journal of clinical investigation 123(1): 517–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Murakami R, Matsumura N, Mandai M, Yoshihara K, Tanabe H, Nakai H, Yamanoi K, Abiko K, Yoshioka Y, Hamanishi J, Yamaguchi K, Baba T, Koshiyama M, Enomoto T, Okamoto A, Murphy SK, Mori S, Mikami Y, Minamiguchi S and Konishi I (2016). "Establishment of a Novel Histopathological Classification of High-Grade Serous Ovarian Carcinoma Correlated with Prognostically Distinct Gene Expression Subtypes." The American journal of pathology 186(5): 1103–1113 [DOI] [PubMed] [Google Scholar]
- 24.Vargas HA, Huang EP, Lakhman Y, Ippolito JE, Bhosale P, Mellnick V, Shinagare AB, Anello M, Kirby J, Fevrier-Sullivan B, Freymann J, Jaffe CC and Sala E (2017). "Radiogenomics of High-Grade Serous Ovarian Cancer: Multireader Multi-Institutional Study from the Cancer Genome Atlas Ovarian Cancer Imaging Research Group." Radiology 285(2): 482–492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Vargas HA, Micco M, Hong SI, Goldman DA, Dao F, Weigelt B, Soslow RA, Hricak H, Levine DA and Sala E (2015). "Association between morphologic CT imaging traits and prognostically relevant gene signatures in women with high-grade serous ovarian cancer: a hypothesis-generating study." Radiology 274(3): 742–751 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ohsuga T, Yamaguchi K, Kido A, Murakami R, Abiko K, Hamanishi J, Kondoh E, Baba T, Konishi I and Matsumura N (2017). "Distinct preoperative clinical features predict four histopathological subtypes of high-grade serous carcinoma of the ovary, fallopian tube, and peritoneum." BMC cancer 17(1): 580. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1: Among 80 patients with adnexal sampling site, CT-molecular subtype had independent predictive values after adjusting for age, ASA score (=3), stage (=IV).
Supplementary Figure 2: Multivariate prediction of high surgical complexity among optimal debulking patients with primary tumors as molecular subtype tissue sources (n=80)