e18166 Background: Watson for Oncology (WFO) is a Memorial Sloan Kettering trained cognitive computing system designed to extract structured and unstructured data from medical records; ingest a large body of published evidence and guidelines; and provide evidence-based treatment options tailored to an individual patient. In Mexico, 160,000 new cases of cancer are diagnosed per year, causing almost 80,000 deaths. 70% of cases are diagnosed at a late stage (Secretariat of Health, 2016). Given the burden of cancer on morbidity and mortality in Mexico, WFO was evaluated as a potential tool to help oncologists select treatment options for their patients. Our objective was to gain feedback on the experience of using WFO in a Mexican population by a panel of local oncology experts, retrospectively comparing their treatment recommendations to the options provided by WFO. Methods: Patient cases with lung, breast, gastric, colon and rectal cancer diagnosed within the last 5 years were included. Physician users were asked to describe the perceived utility of WFO and to compare the treatment options proposed by WFO with local treatment decisions and guidelines. Results: A total of 100 cases were processed using WFO. Oncologists found WFO to be very useful. They felt that it would be particularly beneficial in clinics that lack subspecialist expertise, and for training medical students and residents. Differences between WFO recommendations and those of the Mexican oncologists were attributed primarily to limited access to high-cost therapies in Mexico and differences between U.S. and Mexican cancer treatment guidelines. Conclusions: WFO is a cognitive computing tool that has the potential to improve the consistency, efficiency, and overall quality of cancer care in Mexico. By surfacing relevant literature and evidence, WFO may provide support for coverage of effective yet high-cost therapies.

PTP1B is involved in the oncogenesis of breast cancer. In addition, neoadjuvant therapy has been widely used in breast cancer; thus, a measurement to assess survival improvement could be pathological complete response (pCR). Our objective was to associate PTP1B overexpression with outcomes of breast cancer patients who underwent neoadjuvant chemotherapy. Forty-six specimens were included. Diagnostic biopsies were immunostained using anti-PTP1B antibody. Expression was categorized as negative (<5%) and overexpression (≥5%). Patients’ responses were graded according to the Miller–Payne system. Sixty-three percent of patients overexpressed PTP1B. There was no significant association between PTP1B overexpression and pCR (P = 0.2). However, when associated with intrinsic subtypes, overexpression was higher in human epidermal growth factor receptor 2-positive-enriched specimens (P = 0.02). Ten-year progression-free survival showed no differences. Our preliminary results do not show an association between PTP1B over-expression and pCR; however, given the limited sample and heterogeneous treatment in our cohort, this hypothesis cannot be excluded.

Background: In Mexico 60% of newly diagnosed BC patients present with locally advanced disease. In the United States, 88-95% of patients present with early stage disease, however in Mexico, only 10% of patients are diagnosed at this stage. The Seguro Popular (SP), a national insurance scheme for Mexico's poor and uninsured population, was implemented for BC in 2007. The SP covers diagnostics, local and systemic treatment for BC patients. The INCMNSZ is a third level hospital where patients receive diagnosis and treatment for non-malignant diseases, and also routinely conducts BC annual mammography screening for women ≥ 50 years old. We aimed to evaluate BC outcomes in our patients following the implementation of SP at our institution. Methods: From January 2011 to May 2013, we prospectively evaluated a cohort of 93 patients with BC included in the SP program at the INCMNSZ. A retrospective cohort of 99 patients without SP access who received neoadjuvant chemotherapy from January 2007 to December 2009 was used for comparative analysis. Complete pathologic response (pCR) was defined as complete absence of invasive tumor in breast and axillary nodes. We used Fisher's exact test to compare pCR between groups. Results: A total of 93 patients were diagnosed with BC and included in the SP program. 10 patients were excluded from the analysis- (5 were lost to follow-up and 5 had missing data. Median age was 52 (46-59) years and 37 patients (45%) were premenopausal at diagnosis), median follow up 11.8 months. 20 patients (24%) had stage I, 36 (43%) stage II, 7 (8%) stage III, and 20 (24%) stage IV. 55 (66%) patients had hormone receptor (HR) positive (ER + and/or PR+), 15 (18%) HER2+ and 13 (16%) had triple negative (TN) tumors. Thirty-two (39%) patients received neoadjuvant chemotherapy. 27% of patients achieved pCR (17% in HR positive, 20% in TN and 50% in HER2+ tumors) compared to 15% in the non SP cohort (11% in HR positive, 22% in TN and 14% in HER2+ tumors). HER2+ tumors from the SP cohort reached significantly higher pCR rates than in the non SP cohort (p = 0.029). Discussion: Our results confirmed that about half of our patients were diagnosed with early stage BC, which highlights the importance of a screening program at our institution. Increased pCRs in the HER2+ subgroup may reflect wider access to trastuzumab after implementation of SP. Although we acknowledge a short period of follow-up in our cohort, our results in pCR are comparable to those previously reported in the literature. We believe the SP had a positive impact on access to care and outcomes in our patients. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-09-08.

e18566 Background: Chronic myelomonocytic leukemia (CMML) is the most aggressive of chronic leukemias, with a short overall survival and a high transformation rate to acute leukemia. We investigated factors related to blastic transformation in Mexican population treated in a tertiary referral center Methods: Records of patients diagnosed with CMML between 2000-2015 were reviewed; patients with incomplete data were excluded. IBM SPSS Statics 21.0 software was used to perform statistical analysis. Results: 54 patients were included, with a median age of 71 years and an overall survival of 16 months. The rate of blastic transformation found was 33% (18 patients), with a time to progression of 9 (0-87) months. Comparing patients who didn’t underwent blastic transformation to those who did, those who progress to acute leukemia tend to be younger (58 vs. 71 years, p = 0.001), have a higher peripheral blood blast count (2% vs. 0%, p = 0.003), where more likely to have immature myeloid precursors circulating in peripheral blood (94% vs. 64%, p = 0.02). In multivariate analysis, age continued to be statistically significant (HR 0.97, 95% IC = 0.929-0.987). There where no statistical difference in the two groups regarding hemoglobin levels (8.2g/dL vs. 10.1g/dL) platelets count (115 X 109 vs. 93 X 109), absolute neutrophil count (5.83 X 109 vs. 5.18 X 109), absolute monocyte count (3.09 X 109 vs. 2.680 X 109), and bone marrow blast count (0 vs. 2%). Cytogenetic considered as high risk was not predictor of blastic progression. Intensive chemotherapy was offered to 7(38.9%) patients, with a complete response rate of 0%, the overall survival was 1.4 months (0-9). Conclusions: Contrary to other reported series;Mexican patients with CMML that progresses to overt acute leukemia were considerably younger, with a higher tumor burden and a very short overall survival. In this population, it is important to consider more aggressive treatments at diagnosis, focusing in high dose chemotherapy and hematopoietic stem cell transplantation in a short term ratter than watching and waiting or using agents that do not impact in disease natural history.

e15567 Background: Primary debulking surgery is considered the standard of treatment in advanced ovarian cancer (AOC) while neoadjuvant chemotherapy is used in non resectable stages. Methods: We retrospectively analyzed 68 AOC cases from January 2000 to December 2011. 35 received neoadjuvant chemotherapy (NAC) while 33 were resected primarily and received adjuvant chemotherapy (AC). To compare both groups we used T standard test and 2 sided chi -squared. Non parametric variables were analyzed with U Mann – Whitney. Overall survival(OS) was analyzed using Kaplan Meier method with log rank test. We considered statistically significant p<0.05. (SPSS v 17). Results: The median age was 60 (NAC) vs 53(AC) years respectively. The NAC group had more advanced stage disease (FIGO IIIC/IV stage; 71/29% vs 91/9%; p=0.04). The most frequent histologic subtype in both groups was serous - papillary and histologic grade was poorly differentiated in 71 vs 72% (p=0.41). At diagnosis the median levels of ca-125 were 1,896 U/ml for NAC group vs 768 U/ml for AC group (p=0.025). After the primary treatment received the median levels of ca – 125 were 29 U/ml vs 84 U/ml (p=0.76). Platin based chemotherapy was used in 95% vs 70% respectively. Complete resection of macroscopic disease was observed in 68% of NAC vs 63% in AC respectively. No statistical differences were observed in surgical time (median 192 min vs 204 min; p=0.55) and surgical bleeding (468 vs 510ml; p=0.79). Median survival time was 23 +/-26 months for neoadjuvant and 27+/- 35 months for primary surgery (p=0.56). In a subgroup analysis of patients who received 6 neoadjuvant cycles vs perioperatory chemotherapy (3 pre and 3 postoperatory) we observed a significant survival difference, with a median of 62 months (95% CI 12-38) vs 12 months (95% CI 54 – 178) respectively (p=0.010). Conclusions: Non inferior survival or differences in surgical outcomes were observed with neoadjuvant therapy. Significant survival increase was observed in patients who received complete chemotherapy schedule before surgery, this evidence allowed to design a prospective trial in our Institution.