Search

This study was undertaken with the hypothesis that certain common morphologic features of ovarian carcinomas are predictably associated with BRCA1 and BRCA2 deficiencies. We selected 43 high-grade serous carcinomas diagnosed at Memorial Sloan-Kettering Cancer Center that were studied as part of The Cancer Genome Atlas pilot project. In addition to 12 randomly selected nonfamilial BRCA-unassociated cases, all 31 Memorial Sloan-Kettering Cancer Center cases with BRCA1 or BRCA2 abnormality were included (n=43). Slides were examined to assess tumor architecture, mitotic index, tumor-infiltrating lymphocytes (TILs), nuclear pleomorphism, necrosis, and involvement of fallopian tube epithelium. Comparing BRCA1-associated cases (BRCA1 germline mutation, n=4, BRCA1 somatic mutation, n=6, BRCA1 promoter methylation, n=13) with unassociated cases (n=12) identified statistically significant differences in morphology. BRCA1-associated high-grade serous carcinomas had more frequent Solid, pseudoEndometrioid, and Transitional cell carcinoma-like morphology (SET features) (P=0.0045), higher mitotic indexes (P=0.012), more TILs (P=0.034), and either geographic or comedo necrosis (P=0.034). BRCA2-associated cases (germline mutation, n=4 and somatic mutation, n=4) tended to show SET features, but they were relatively deficient in TILs and necrosis. Two algorithms incorporating tumor architecture, necrosis, and either mitotic indexes or TILs separated cases that showed 2 of 3 features (BRCA1 associated) from those with 0 of 3 features (BRCA unassociated; P=0.0016 and P=0.0033). A test set comprising 9 BRCA1 germline mutants and 14 high-grade serous carcinoma controls lacking BRCA1 and BRCA2 germline mutation was used to validate the algorithms, with specific emphasis on the ability to detect cases with BRCA1 germline mutation. Best results were obtained with the algorithm that incorporated SET features, necrosis, and mitotic index (P=0.0072; sensitivity of 1.0 (95% CI, 0.66-1.0); specificity of 0.57 (95% CI, 0.29-0.82); positive predictive value of 0.60 (95% CI, 0.32-0.84) and a negative predictive value of 1.0 (95% CI, 0.63-1.0)). These preliminary data indicate potential strong associations between morphology and genotype in high-grade serous carcinomas.

Patients with hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome are prone to develop smooth muscle tumors of the uterus and skin and also renal carcinomas. The morphologic features of renal tumors that arise in the setting of HLRCC are well described, the hallmark feature being the presence of prominent eosinophilic nucleoli surrounded by a clear halo. Renal tumors associated with HLRCC are aggressive and often present late with high-stage disease. Early detection of patients with HLRCC could lead to surveillance for renal carcinomas. Women with HLRCC often present at a young age with uterine leiomyomas and frequently undergo hysterectomy as a result. HLRCC associated uterine leiomyomas show nuclear features similar to those described in HLRCC associated renal tumors. Therefore, presence of these nuclear features in uterine smooth muscle tumors may aid in early detection of HLRCC, resulting in surveillance for renal carcinoma. We present a case of a 32-year-old woman who underwent hysterectomy for uterine fibroids and subsequently presented 5 years later with high-stage renal carcinoma. Histologic evaluation of uterine leiomyomas and renal carcinoma revealed identical nuclear features, that is, prominent nucleoli with perinucleolar halos, raising the possibility of HLRCC. Subsequent testing confirmed a fumarate hydratase mutation. The presence of above-described nuclear features in uterine leiomyomas should raise the possibility of HLRCC.

The objective of this study was to determine the incidence of invasive cancer in ovarian masses diagnosed as borderline tumor (BT) at the time of frozen section.We performed a retrospective review of all patients diagnosed with ovarian BT on frozen section (FS) at our institution between 2000 and 2010. Clinical and pathologic data were extracted. Univariate and multivariate analyses were performed using standard two-sided statistical tests.A total of 120 patients were identified, of which 104 (86.7%) had BT on frozen section that was confirmed on final pathology. In 15 (12.5%) patients, BT was diagnosed on FS but was reclassified as invasive cancer on final pathology. One patient (0.8%) had BT on FS but benign pathology on final diagnosis. Histologies included serous in 79 (65.8%), seromucinous in 13 (10.8%), mucinous in 21 (17.5), endometrioid in 5 (4.2%), and clear cell in 2 (1.7%) patients. Reclassification of pathologic diagnosis was related to histologic subtype, but only for endometrioid and clear cell tumors (P0.001). The rate of invasive cancer for serous micropapillary tumors on frozen section was 42.8% compared with 2.8% for serous non-micropapillary tumors (P0.001). Tumor size8 cm was associated with a 22.4% incidence of invasive cancer on final pathology compared to 3.2% in tumors ≤ 8 cm (P=0.004).Comprehensive surgical staging can be considered in BT8 cm in diameter, as well as those with micropapillary serous, endometrioid, and clear cell histology diagnosed at the time of frozen section analysis.

The hereditary basis of endometrial cancer is apparent in young women with endometrial cancer. The objective of this study was to examine risk factors and outcomes in patients 40 years of age and younger with endometrial cancer.We performed a retrospective cohort study of patients aged 40 years or less who were diagnosed with endometrial carcinoma between 1/93 and 5/08. Clinical and pathologic data were extracted from medical records. Paraffin-embedded slides from hysterectomy specimens were obtained and DNA mismatch repair (MMR) immunohistochemistry was performed. Cases were analyzed according to the presence of DNA MMR protein defects. Standard two-sided statistical tests were performed.Of the 56 identified patients, the median age was 36 years (range, 24-40). The majority of the endometrial carcinomas were of endometrioid histology (91%), stage I (71%), and FIGO grade 1 (59%). Abnormal DNA MMR was found in 9 cases (16%). Cases with abnormal DNA MMR had lower body mass index (BMI) (P=0.05), and had a family history suggestive of Lynch syndrome (P=0.001). Tumors were more likely to have advanced stage disease (P0.001), be high grade (P0.001), have deep myometrial invasion (P0.001), and have lymphovascular invasion (P=0.002). Cases with abnormal DNA MMR had significantly worse overall survival (P=0.028) and progression-free survival (P=0.042).Endometrial cancer is rare in women aged 40 years or less. In this group of patients, loss of DNA MMR was associated with lower BMI, worse clinicopathologic factors, and worse outcome. These results may have implications when young women diagnosed with endometrial cancer are counseled regarding prognosis.

Objective To determine the incidence of adnexal and lymph node (LN) metastasis in newly diagnosed endometrial stromal sarcoma (ESS). Methods We identified all cases with a diagnosis of ESS evaluated at our institution from January 1, 1980 to October 31, 2009. All uterine pathology was reviewed at our center. High-grade or undifferentiated tumors and ESS arising in extrauterine sites were excluded. Pertinent clinical data were abstracted from electronic medical records. Appropriate statistical tests were performed using SPSS16.0. Results We identified 94 cases of ESS. LN metastasis was identified in 7 (19%) of 36 patients who underwent LN evaluation. Six of the 7 cases with LN metastasis had lymphovascular invasion (LVI). LVI status was not reported in the other case. Five of the 7 patients with LN metastasis had grossly positive LNs with or without other gross extrauterine disease. Of 20 patients with disease grossly limited to the uterus and grossly normal LNs, 2 (10%) had LN metastasis. Both of these cases had LVI and extensive myoinvasion. Eighty-seven cases (93%) underwent salpingo-oophorectomy. Adnexal metastasis was identified in 11 (13%) of 87 cases, all manifested by gross adnexal tumor and occurring in patients with other gross pelvic extrauterine disease. Conclusion The incidence of LN metastasis in ESS is commonly associated with gross extrauterine disease, extensive myoinvasion, and LVI. Since myoinvasion and LVI status often are not assessable at the time of hysterectomy, LN dissection remains a reasonable option at primary surgery. The rate of adnexal metastasis appears to be negligible in the absence of gross adnexal and extrauterine tumor.

5043 Background: The optimal use of BRCA1 germline testing in patients with HGS-OC is unclear. Moreover, BRCA1 germline testing does not provide information on non-germline mechanisms of BRCA1 loss. We investigated the performance of BRCA1 immunohistochemistry (IHC) testing in HGS-OC with known BRCA1 mutation status. Methods: Eligible patients had HGS-OC, underwent surgery at a single institution between 1997 and 2010, and were tested for germline BRCA1 mutations under IRB-approved protocols. FFPE tumor tissue was used in triplicate to construct a tissue microarray (TMA). Two pathologists, blinded to BRCA1 status, independently scored each sample as BRCA1 IHC present (normal) or absent (abnormal). Whole sections were stained for all samples with absent BRCA1 staining on TMA. A commercially available monoclonal antibody against BRCA1, clone MS110 from Calbiochem (OP92) was used with previously optimized conditions. Results: 123 samples (30 BRCA1 +; 93 BRCA1 wild-type) were analyzed and 47 (38%) had abnormal BRCA1 IHC. Inter-observer agreement on BRCA IHC was high (kappa=0.87). BRCA IHC had a sensitivity of 83.3% (CI: 70.0-96.7%), specificity of 76.3% (CI: 67.7-85.0%), PPV of 53.2% (CI: 38.9-67.5%), and NPV of 93.4% (CI: 87.9-99.0%) for BRCA1 germline mutation. There was a trend towards improved overall survival in the BRCA IHC abnormal vs. normal patients (median survival 82 vs 61 wks, HR 0.70, p=0.12). Conclusions: BRCA1 IHC is a reproducible and inexpensive test with a high NPV for absence of a BRCA1 germline mutation. The 22 (17.7%) cases with abnormal BRCA1 IHC and wild-type BRCA1 germline status may have other mechanisms of protein loss such as promoter hypermethylation or somatic mutation and are currently being tested for these changes. This rate of non-germline BRCA1 loss in HGS-OC is consistent with data generated by the Tumor Cancer Genome Atlas Network. BRCA1 IHC may be useful as a reliable biomarker for risk stratification in HGS-OC. [Table: see text]

5030 Background: LGS OC develops in a step-wise pattern from serous cystadenoma to SB tumor to invasive LGSOC. LGSOC accounts for 6-22% of serous OC and is a chemoresistant disease. Prior studies have reported that LGS/SB ovarian tumors harbor BRAF mutations in 28-35% of cases, suggesting that LGS/SB OC may respond to mutant BRAF inhibitors. Methods: We genotyped 75 LGS and SB ovarian tumors for BRAF and KRAS hotspot mutations using a mass spectrometry based Sequenom assay. All samples were collected at MSKCC between 2000-2010. All specimens underwent two independent pathologic reviews for diagnostic confirmation and macrodissection to ensure 80% cellularity. The incidence and identity of BRAF and KRAS mutations were defined and results were correlated with stage, response to treatment, and outcome. Exon capture sequencing is underway to sequence all coding exons of 230 cancer associated genes to identify additional targetable alterations in LGS/SB OC. Results: Of 75 samples examined, 56(75%) were SB and 19(25%) LGS histology. 57% of samples showed KRAS mutation (KRAS+) (n=17, 23%) or BRAF mutation (BRAF+) (n=26, 35%). Mutation status was mutually exclusive. All BRAF + were V600E, KRAS+ were G12D (n=9) or G12V (n=8). BRAF+ was associated with early stage (I-II) at presentation and SB histology. 22 (29%) patients were treated with chemotherapy, 2 (9%) with KRAS + and 0 with BRAF+. 4(15%) BRAF+ patients underwent resection of recurrent disease. All BRAF+ patients are alive without evidence of disease at a median follow-up of 43.3 months (range 1.9- 129.3 months). Conclusions: V600E BRAF mutations are present in 35% of cases of SB/LGS ovarian cancer. Presence of BRAF + in SB/LGS OC is associated with good prognosis and surgical cure of disease. Patients with SB/LGS OC who require systemic therapy are unlikely to have BRAF+. Our group is performing a detailed search for additional targetable alterations in multiple signal transduction pathways through next generation sequencing technology. [Table: see text]

Objectives: The PI3K/AKT pathway is frequently deregulated in endometrial cancer due to somatic mutations and other molecular mechanisms. New drugs directed toward these pathways are currently in clinical trials. Many patients have available primary tumor specimens for correlative studies or enrollment eligibility. It is unknown if the same genomic alterations found in primary disease are also present in recurrent specimens. We hypothesize the genomic landscapes of primary and recurrent endometrial cancers are similar. Methods: We identified 18 patients with available paired primary and recurrent endometrial cancer specimens. Histologic diagnoses were confirmed and categorized based on current FIGO criteria. DNA was extracted from microdissected FFPE tissue sections. Mass spectrometry mutation detection was used to assess 40 hot spot mutations in AKT1, KRAS, NRAS, and PIK3CA. Sanger sequencing was used to identify mutations in PTEN. PTEN IHC was performed using validated antibodies. Results: Specimens included 9 cases with endometrioid histology, 8 with serous histology and 1 with mixed serous-endometrioid histology. Five patients had specimens from 2 separate recurrences. Of the 18 primary tumors, 4 (22%) had PIK3CA mutations, 5 (28%) had PTEN mutations, and 1 (5.5%) had a KRAS mutation. One PIK3CA and two PTEN mutations were each present in a primary tumor but not identified in the paired recurrent specimen. One NRAS mutation was identified in a secondary recurrent specimen but not found in either the paired primary or initial recurrent specimen. One AKT1 mutation was identified in both an initial and secondary recurrent specimen but not found in the corresponding primary specimen. PTEN immunostaining was absent in 9 (50%) primary specimens and in one of these cases PTEN staining found to be present in the recurrence. Conclusions: As clinical trials increasingly incorporate molecularly targeted agents, mutational and other genomic profiling of tumors will be helpful to identify patients most likely to benefit from these drugs. These data suggest that most recurrent endometrial carcinomas harbor similar genomic alterations in the PI3K/AKT pathway as their primary counterparts. Further studies will be required to expand upon these preliminary findings and determine if repeat biopsy of recurrent disease in clinical trial participants is necessary. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4168. doi:10.1158/1538-7445.AM2011-4168

5589 Background: To determine the incidence of adnexal and lymph node (LN) metastases in endometrial stromal sarcoma (ESS). Methods: We identified all cases evaluated at our institution with a diagnosis of ESS from January 1, 1980 to December 31, 2007. All uterine pathology was reviewed at our center. High-grade or undifferentiated tumors and ESS arising in extrauterine sites were excluded. Pertinent clinical data were abstracted from electronic medical records. Appropriate statistical tests were performed using SPSS 15.0. Results: We identified 91 cases with ESS. All except two underwent hysterectomy. Eighty-four cases (92%) underwent salpingo-oophorectomy (bilateral in 81 [96%]). Adnexal metastases were identified in 8 (9.5%) of 84 cases. All adnexal metastases were manifested by gross adnexal tumor and occurred in patients with other gross pelvic extrauterine disease. Seven of these 8 patients also had lymph-vascular space invasion (LVSI). LVSI status was not reported in the other patient. In patients with gross pelvic extrauterine disease, adnexal metastasis did not affect survival. LN metastases were identified in 8 (24%) of the 34 patients who underwent LN evaluation. Thirty-three evaluations were performed intraoperatively and one was via postoperative CT-guided biopsy of an enlarged LN. Five of the 8 cases with LN metastases had LVSI. LVSI status was not reported in the other 3 cases. Of 20 patients with disease grossly limited to the uterus and grossly normal LN, 2 (10%) had LN metastases. Both of these cases had LVSI and extensive myoinvasion approaching the serosa. Four of the remaining 6 patients with LN metastases had grossly positive LN and other gross extrauterine disease. The final 2 patients had grossly positive LN without other extrauterine disease. Conclusions: The incidence of LN metastases in ESS is most commonly associated with gross extrauterine disease, extensive myoinvasion and LVSI. However, since myoinvasion and LVSI status often are not assessable at the time of initial hysterectomy, LN dissection remains a reasonable option at primary surgery. The rate of adnexal metastasis appears to be negligible in the absence of gross adnexal and extrauterine tumor, but there may be a role for oophorectomy in the hormonal management of ESS. No significant financial relationships to disclose.

5579 Background: The significance of the hereditary basis of endometrial cancer is apparent in young women with endometrial cancer. The objective of this study was to describe the incidence and prognostic implications of DNA mismatch repair protein defects in a series of patients age 40 years and younger with endometrial cancer. Methods: We performed a retrospective cohort study of patients age 40 years or less who were diagnosed with endometrial carcinoma between 1/93 and 5/08. Clinical and pathologic data were extracted from medical records. Paraffin-embedded slides from hysterectomy specimens were obtained and DNA mismatch repair (MMR) immunohistochemistry (IHC) was performed. Cases were analyzed according to presence of DNA MMR protein defects. Standard two-sided statistical tests were performed. Results: Of the 71 identified patients, the median age was 37 years (range, 24–40), with a median follow-up of 47 months (range, 1–178). The majority were of endometrioid histology (94%), stage I (73%), and FIGO grade 1 (61%). IHC was performed with available blocks (n = 56), and loss of DNA MMR was found in 9 cases (16%). Cases with loss of DNA MMR were more likely to have high-grade (FIGO 2 or 3) tumors (p < 0.001), be advanced stage (III or IV) at the time of diagnosis (p < 0.001), and have a family history suggestive of hereditary non-polyposis colorectal cancer (p = 0.01). There was no difference between the groups in histology, obesity (BMI>30), parity, or history of infertility. Analysis of clinical outcomes revealed that cases with loss of DNA MMR had significantly worse overall survival (median survival not reached, log rank p = 0.018). At the time of last follow-up, 2 (22%) patients with loss of DNA MMR were dead of disease compared with 2 (4%) patients with retained DNA MMR. Conclusions: Endometrial cancer is rare in young women age 40 years or less. In these patients, loss of DNA MMR was associated with worse clinicopathologic factors (high-grade and advanced-stage tumors) and worse outcome. Other clinical risk factors associated with endometrial cancer were not associated with loss of DNA MMR. The data suggest that MMR testing in women 40 years of age or younger with endometrial cancer may have clinically useful prognostic information. No significant financial relationships to disclose.