Search

Introduction: Classification by intrinsic subtypes by gene expression profiles of early-stage breast cancer (EBC) provides information of prognostic value and constitutes a tool to help in making therapeutic decisions. Several authors have proposed surrogated classifications based on immunohistochemistry results (IHC) in order to facilitate a classification with identical prognostic and predictive value. However, there is evidence that suggests a lack of correlation between these classifications. The aim of this study was to evaluate the correlation between classification by intrinsic subtypes in patients with EBC ER+/ HER2neg, obtained by PAM50 and the surrogated classification proposed by St. Gallen 2013. Methods: Samples from 12 centers from the spanish region of Castilla y León were analyzed by PAM50 (nCounterTM / Nanostring) at the University Hospital of Salamanca. The results obtained were compared with the surrogate classification of St.Gallen'13 from local reports. Tumors of patients pre and post-menopausal tests T1-2, N0-N1mi, grade I-II that met criteria for inclusion of the regional evaluation program through ProsignaTM were included. Results: Between August'15 and December'17, 264 samples were analyzed. All patients were classified by IHC as lum-A or lum-B. In total 113 cases were reclassified by PAM50 (43%). The change of LumA by IHC to Lum-B by PAM50 was 18%, while Lum-B by IHC to Lum-A by PAM50 was 58% (n = 89). In those cases considered Lum-B by IHC based only in a value of Ki67>14% (n = 93), 54% was reclassified to Lum-A. Conversely, when low expression (negative or Conclusion: The surrogated classification by IHC of intrinsic subtypes in EBC ER+/HER2neg shows a low concordance with PAM50 analysis, and cannot be considered adequate. In particular, the presence of negative or Citation Format: Rodriguez CA, Guillen C, Garcia M, Sancho M, Gomez A, Ludena D, Cruz JJ. Discordance between surrogated intrinsic subtypes defined by immunohistochemistry compared with PAM50 in ER positive / HER2 negative early breast cancer. Analysis of value of the status of the progesterone receptor and Ki67 [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-08-07.

4022 Background: Sequential treatment with nab-paclitaxel, gemcitabine followed by oxaliplatin, leucovorin and 5-fluorouracil ( nab-P/Gem-mFOLFOX) have shown a good safety profile and clinical activity in metastatic pancreatic ductal adenocarcinoma (mPDAC) in a previously published phase I SEQUENCE trial. Methods: We have compared nab-P/Gem-mFOLFOX to the standard nab-P/Gem in first-line treatment, in an open-label multi-institutional prospective randomised phase II trial in patients with untreated mPDAC from 14 Spanish hospitals. Patients were allocated 1:1 to nab-paclitaxel (125 mg/m2) plus gemcitabine (1000 mg/m2) on days 1, 8 and 15, followed by modified FOLFOX-6 (oxaliplatin [85 mg/m2], L-leucovorin [200 mg/m2] or racemic leucovorin [400 mg/m2], 5-fluorouracil bolus [400 mg/m2], and 5-fluorouracil 48-hour continuous infusion [2400 mg/m2]) on day 29 of a 6-week cycle or nab-paclitaxel (125 mg/m2) plus gemcitabine (1000 mg/m2) on days 1, 8 and 15 of a 4-week cycle. The primary endpoint was the 12-month overall survival rate (OS) in randomised patients. EudraCT number 2014-005350-19; ClinicalTrial.gov identifier NCT02504333. Results: Between July 27, 2017, and April 16, 2019, 182 patients were screened and 157 randomised: 78 to nab-P/Gem-mFOLFOX and 79 to nab-P/Gem. Patients receiving nab-P/Gem-mFOLFOX showed a significantly higher 12-month, and 24-month OS (95% CI): 55.3% (44.2-66.5%) versus 35.4% (24.9-46.0%) (p = 0.016), and 22.4% (13.0-31.8%) versus 7.6% (1.8-13.4%) (p = 0.012), respectively. The median OS (95% CI) reached 13.2 (10.1-16.2) months with nab-P/Gem-mFOLFOX and 9.7 (7.5-12.0) months with nab-P/Gem (HR = 0.676, 95% CI 0.483-0.947, p = 0.023). 39.7% patients in the nab-P/Gem-mFOLFOX group and 54.4% in nab-P/Gem group received subsequent anticancer treatments. Safety was comparable except for grade ≥3 neutropenia (46.1% versus 24.1%, p = 0.004) and grade ≥3 thrombocytopenia (23.7% versus 7.6%, p = 0.007) that were higher in the nab-P/Gem-mFOLFOX regimen. Two (2.6%) patients died due to adverse events in the nab-P/Gem-mFOLFOX arm. Conclusions: Nab-P/Gem-mFOLFOX showed significantly higher clinical activity than the standard nab-P/Gem treatment, with a manageable safety profile. This regimen represents a feasible and efficient new option for first-line treatment of mPDAC. Clinical trial information: NCT02504333.

Background Immunohistochemistry (IHQ) is commonly used for the detection of mismatch repair proteins deficiency (MMRD). One very infrequent abnormal pattern of MMR protein expression is the loss of PMS2 and MSH6, with intact expression of MLH1 and MSH2. Case presentation We review the frequency of this MMRD IHC pattern among 108 colorectal (CRCs) and 35 endometrial cancers in our files with loss of expression of at least one protein, and present two CRCs showing loss of PMS2 and MSH6 protein expression (1.9% of CRCs). NGS analysis of these tumours identified PMS2 mutations (R134* germline mutation in one tumour and M1R and c.1239delA somatic mutation in the other) as the primary event and somatic MSH6 mutation (c.3261dupC) as the secondary event in both tumours. Conclusions This study suggests that Next Generation Sequencing (NGS) tumour analysis should be considered in the algorithm of Lynch syndrome screening to detect MMR gen somatic mutation in inconclusive cases.

3507 Background: FOLFOXIRI+BEV has demonstrated a survival benefit compared with FOLFIRI plus BEV (TRIBE Lancet Oncol 2015) in first-line mCRC. Nevertheless, due to its safety profile, this schedule is not recommended for all pts. In addition, we have showed that the detection of ≥3 bCTCs is a poor prognostic factor for survival (MACRO The Oncologist 2012). The VISNU-1 trial compares FOLFOX + BEV vs FOLFOXIRI + BEV in pts with mCRC and ≥3 bCTCs. Progression-free survival (PFS) is the primary endpoint. Secondary endpoints included overall response rate (ORR) and overall survival (OS). Methods: This is an open, multicentric, randomized phase III trial. Patients with mCRC younger than 70 years, ECOG 0-1 were randomized to FOLFOX+BEV (arm A) or FOLFOXIRI+BEV (arm B), stratified per KRAS mutation (mutated vs WT) and number of involved organs (1 vs >1). Results: 349 pts were included in the ITT population; 177 in group A and 172 in group B. Characteristics of the pts, molecular profiling and safety analysis have been previously presented at ASCO 2018 and showed that this schedule had an acceptable toxicity profile. Efficacy analysis in the ITT population is shown in the table. Conclusions: In this population with very bad prognosis, our study met its primary endpoint. Pts who received FOLFOXIRI + Bev benefit for a statistically significative PFS and ORR. OS showed a trend of benefit in the experimental arm. According to these results, FOLFOXIRI-Bev could be considered an adequate treatment option for pts with mCRC and ≥3 bCTCs. Clinical trial information: 2012-000846-37. [Table: see text]

655 Background: Adyuvant chemotherapy is a unified therapeutic principle in colon cancer (CC). However, many prognostic factors are arising to determine who really benefits from adyuvant chemotherapy in order to avoid overtreatment. Primary tumor location of CC is emerging as an important prognostic factor owing to distinct biological features. However, there is hardly any evidence showing the benefit of treatment according to primary tumor location. Methods: We retrospectively included patients with stage II and III CC that underwent surgical resection between 2009 and 2014 HURyC. We performed a multivariable Cox model analysis to estimate the benefit of chemotherapy according to tumor location in terms of DFS and OS. The model was further adjusted by including the following confounders: ECOG-PS, number of removed nodes ( < 12 or ≥ 12), use of adyuvant chemotherapy and age. A covariate was considered a confounder factor if the difference between the adjusted and unadjusted coefficient of chemotherapy varied > 10%. Stata 13.1 was used to analyze the data. Results: 564 patients were identified (267 left sided and 297 right sided). The median follow-up of the entire cohort was 49 months. Globally, chemotherapy did not improve DFS neither in right sided or left sided CC (HR: 0.81, p: 0.58; HR 1.31, p: 0.4, respectively). As for OS, it was improved when adding chemotherapy in both sides (right sided CC HR 0.51, p: 0.061 and left sided CC HR 0.42, p: 0.009). By stages, chemotherapy did not improve DFS or OS according to tumor localization in stage II. In right sided stage III, there was a trend to improve both DFS and OS (HR: 0.54, p: 0.14; HR: 0.39, p: 0.1, respectively). In left sided stage III, there was a trend to improve DFS (HR: 0.66, p: 0.36) and there was an improvement in OS (HR: 0.23, p < 0.001). Conclusions: The multivariable analysis showed benefit of chemotherapy in both right and left sided CC in terms of OS, exhibiting more benefit those patients with left sided CC.

Background The oral multikinase inhibitor regorafenib improves overall survival (OS) in patients with metastatic colorectal cancer (CRC) for which all standard treatments have failed. This study investigated regorafenib plus modified FOLFOX (mFOLFOX6) as first-line treatment of metastatic CRC. Methods In this single-arm, open-label, multicentre, phase II study, patients received mFOLFOX6 on days 1 and 15, and regorafenib 160 mg orally once daily on days 4-10 and 18-24 of each 28-day cycle. The primary end-point was centrally assessed objective response rate (ORR). Secondary end-points included disease control rate (DCR), OS, progression-free survival (PFS) and safety. Results Median overall treatment duration with any study drug was 9.9 months (range 0.6-19.6); median treatment duration with regorafenib was 7.7 months (range 0.1-19.5); six patients remained on regorafenib for more than 1 year. Fifty-three patients received at least one dose of regorafenib. ORR was 43.9% (all partial responses); DCR was 85.4%; median OS was not reached; median PFS was 8.5 months. Treatment-emergent adverse events were experienced by all patients but were manageable with dose modifications. Conclusion Regorafenib + mFOLFOX6 as first-line treatment in patients with metastatic CRC did not improve ORR over historical controls. Regorafenib plus mFOLFOX6 did not appear to be associated with a markedly worse tolerability profile versus mFOLFOX6 alone. ClinicalTrials.gov identifier: NCT01289821.

Background Ductal adenocarcinoma of the pancreas is the fourth leading cause of cancer mortality in the US. The disease is mostly unresectable at the time of diagnosis, making chemotherapy the backbone of therapy. TTFields are a non-invasive, regional antimitotic treatment modality, which has been approved for the treatment of recurrent and newly diagnosed glioblastoma by the FDA. TTFields act by delivering intermediate frequency alternating electric fields to the tumor, and act predominantly by disrupting the formation of the mitotic spindle during metaphase and by dielectrophoretic dislocation of intracellular constituents during cytokinesis. In vitro, TTFields led to reduction in the proliferative potential of pancreatic cancer cell lines with or without chemotherapies, and to a subsequent reduction in their clonogenic potential. In vivo, TTFields led to a significant reduction in tumor volumes. PANOVA was the first trial testing TTFields (150kHz) in pancreatic cancer patients. Results from the first arm of the study, testing TTFields in combination with gemcitabine, have demonstrated superiority in efficacy compared to historical controls, with median PFS of 8.3 months and median OS of 14.9 months (Rivera F. et al, J Clin Oncol 34, 2016 (suppl 4S; abstr 269)). Methods Twenty advanced pancreatic cancer patients were enrolled in the second arm of PANOVA and treated with TTFields in combination with gemcitabine concomitant to nab-paclitaxel. All patients had unresectable tumors, an ECOG performance score of 0-1 and no prior therapy. The primary endpoint was the incidence and severity of treatment emergent adverse events. Results The median age was 68.2 (range – 58-81) and most patients (65%) had an ECOG score of 1. Twelve patients (60%) had distant metastases. The Median number of gem/nab-paclitaxel cycles was 5.0, and median number of TTFields cycles was 7.5. Ten patients (50%) had serious AEs during the study period. Eleven patients (55%) had treatment-related skin toxicity, of which 5 had grade 3 toxicity. No TTFields-related serious AEs were reported. The median PFS was 12.7 months (95% CI 5.4, NA). PFS rate at 6 months was 65%. Of the evaluable tumors, 40% had partial response and another 47% stable disease. The median OS was not reached, and the 1-year survival rate was 72%. Conclusions TTFields concomitant to gemcitabine and nab-paclitaxel are tolerable and safe for advanced pancreatic cancer patients, with promising clinical outcome which doubled published historical data. A phase III trial is planned, testing the efficacy of TTFields combined with gemcitabine and nab-paclitaxel in locally-advanced pancreatic cancer patients. Citation Format: Manuel Benavides, Carmen Guillen, Fernando Rivera, Javier Gallego, Jose Lopez-Martin, Marc Küng. PANOVA: A phase II study of TTFields (150kHz) concomitant with standard chemotherapy for front line therapy of advanced pancreatic adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT130. doi:10.1158/1538-7445.AM2017-CT130

e15790 Background: TTFields are a non-invasive, regional antimitotic treatment modality, which has been approved for the treatment of glioblastoma by the FDA. TTFields predominantly act by disrupting the formation of the mitotic spindle during metaphase. TTFields were effective in multiple preclinical models of pancreatic cancer. PANOVA was the first trial testing TTFields in pancreatic cancer patients. Results from the first arm of the study, testing TTFields in combination with gemcitabine, have demonstrated superiority in efficacy compared to historical controls (Rivera F. et al, J Clin Oncol 34, 2016 (suppl 4S; abstr 269). Methods: Twenty advanced pancreatic cancer patients were enrolled in the second arm of PANOVA and treated with TTFields in combination with gemcitabine concomitant to nab-paclitaxel. All patients had unresectable tumors, an ECOG performance score of 0-1 and no prior therapy. The primary endpoint was the incidence and severity of adverse events. Results: The median age was 68.2 (range – 58-81) and most patients (65%) had an ECOG score of 1. Twelve patients (60%) had distant metastases. Ten patients (50%) had serious AEs during the study period. Eleven patients (55%) had treatment-related skin toxicity, of which 5 had grade 3 toxicity. No TTFields-related serious AEs were reported. The median PFS was 12.7 months (95% CI 5.4, NA): 9.3 months in patients with metastatic disease, and not reached in locally-advanced patients. PFS rate at 6 months was 65%: 50% in metastatic disease and 87.5% in locally advanced patients. Of the evaluable tumors, 40% had partial response and another 47% stable disease. The median OS was not reached, and the 1-year survival rate was 72% (62.5% in metastatic disease and 87.5% in locally advanced disease). Conclusions: TTFields concomitant to gemcitabine and nab-paclitaxel are safe for advanced pancreatic cancer patients, with promising clinical outcome which doubled historical data. A phase III trial is planned, testing the efficacy of TTFields combined with gemcitabine and nab-paclitaxel in locally-advanced pancreatic cancer patients. Clinical trial information: NCT01971281.

790 Background: The role of adjuvant chemotherapy (AC) after chemoradiotherapy and surgery in stage II and III of rectal cancer is not well defined. Neither the regimen nor the duration is clear. Besides, guidelines from different expert groups are conflicting. Methods: 224 patients diagnosed from January 2003 to December 2013 with rectal adenocarcinoma T3/T4 and/or positive node staged by ultrasound or magnetic resonance imaging was collected retrospectively from the Pathology Department data base. Results: Of the 224 patients 61.6% were male, median of age at diagnosis 68.6 years (range 31.3-85.4). All of them received 50.4 Gy with different concomitant regimens (capecitabine, continuous infusion of fluoracil or bolus of fluoracil-leucovorin). 13.4% of patients achieved a pathological complete response. 76.8% of the cohort received adjuvant chemotherapy. The main reasons for not receiving chemotherapy were ECOG ≥ 2, surgery complications and a decision of the patient. 51.2% of patients received an oxaliplatin based regimen and the median time of chemotherapy duration was 4 months (range 1-6 months). Local relapse was diagnosed in 6.3% of patients while a systemic relapse was found in 22.3%. Patients without AC had a mean overall survival (OS) of 85.5 months (CI 95% 70.7-100.4) versus 119.8 monts (CI 95% 110.8-128.8) in the AC cohort (p 0.002) without reaching the median after 58 months (range 8-153.9) median follow-up. Disease free survival (DFS) was also better in the AC cohort with a mean DFS of 104 months (CI 95% 94.3-113.6) versus 71.4 months (CI 95% 55.4-87.4), p = 0.01. Conclusions: AC after chemoradiotherapy and surgery in stage II and III rectal cancer as a standard does not exist. In our cohort, AC showed a significant improve in OS and DFS, but the limitations of a retrospective study should be considered, as well as the impact of surgery on metastasis and salvage chemotherapies. Randomized studies are clearly needed.

751 Background: In the VELOUR trial, adding Z to FOLFIRI improves OS, PFS and RR in mCRC pts progressing after oxaliplatin ±biologic agents. The ASQoP trial (NCT01571284) was designed to gather safety and QoL data from mCRC in real-life setting. We report data collected by the Spanish investigators. Methods: ASQoP is single-arm, open-label trial evaluating safety and QoL of Z in mCRC pts as 2nd line. Eligible pts received Z (4mg/kg) q2wks on day 1/cycle, followed by FOLFIRI (dosing was at physician’s discretion) until disease progression, unacceptable toxicity, death, or investigator/pt decision. The EQ-5D was used for utility index (UI) measure and the EORTC QLQ-C30 as generic cancer instrument. QoL population consisted of pts completing the questionnaire at baseline and ≥1 assessment post-baseline and received ≥1 part of 1 dose of study treatment. Results: The safety population comprised 77 pts with ≥1 completed cycle of treatment. Grade (G)3/4 AEs were reported in 72.7% of pts (vs 83.5% in VELOUR), being G3 most commonly reported. There was no G4 hypertension, stomatitis, or proteinuria. G4 Diarrhea was found in 1.3% of pts. Mean baseline UI was 0.7 (95% CI, 0.63-0.78) in 51 pts, and remained relatively stable at cycles 3 (n=39) and 7 (n=24), with a mean (±SD) change from baseline of 0.03 (±0.26) and -0.06 (±0.35), respectively. Mean baseline global health status score was 63.1 (95% CI, 55.8-70.4) in 54 pts, and remained stable up to cycle 9 with a mean (±SD) change from baseline of 4.17 (±38). Conclusions: Thisanalysis has identified no new safety signals and suggests an acceptable toxicity profile with a relatively stable UI and QoL in Spanish mCRC pts in the real-life setting. [Table: see text]

Pancreatic neuroendocrine tumors (PNETs) are rare malignancies that arise from the islets of Langerhans. The role of standard chemotherapy in advanced well-differentiated PNETs remains to be defined. Sunitinib is an oral multitargeted inhibitor with antiangiogenic and antitumor properties that has shown significant improvement in survival in metastatic PNETs, although objective responses by Response Evaluation Criteria in Solid Tumors were only 9%. We herein report on the case of a middle-aged woman with metastatic PNET who was heavily pretreated for her advanced disease with limited success, and who showed clinical, biochemical, and radiological responses by using Choi criteria but not Response Evaluation Criteria in Solid Tumors criteria. To our knowledge, this is the first reported case of treatment with sunitinib in a patient with PNET in response to Choi criteria.

e21525Background: Given the high risk of toxicity to chemotherapy (CT) in elderly, decision making in this population is complex. In this sense, the project ONCOSARCO aims: 1) To know what (frailty...

4007Background: Hypoxia in PDAC is associated with disease progression and poor prognosis. Evo is a hypoxia-activated prodrug of Br-IPM that is preferentially activated under hypoxic conditions. Th...

e15145Background: Data of outcomes of elderly patients ( ≥ 70) with stage II and III rectal cancer treated with multimodal treatment (neoadjuvant chemoradiotherapy, surgery and adjuvant chemotherap...

269 Background: TTFields are alternating electric fields delivered to the region of the tumor by means of non-invasive transducer arrays. TTFields interfere with mitotic spindle formation, thus having an anti-mitotic activity. In pancreatic cancer, TTFields decreased proliferation and clonogenic potential in vitro, and reduced tumor volume in vivo. The PANOVA trial was designed to test TTFields, combined with chemotherapy, in pancreatic cancer. Methods: Twenty patients with advanced pancreatic adenocarcinoma were enrolled in a prospective, single arm study of TTFields at 150 kHz, concomitant with standard weekly gemcitabine. All patients had histologically-confirmed unresectable tumors, with an ECOG performance score of 0-1 and no prior chemotherapy or radiation therapy. The primary endpoint was incidence and severity of treatment emergent adverse events (AEs). Secondary endpoints included progression free survival (PFS), PFS rate at 6 months, overall survival (OS) and response rate. Results: The median age was 73 (range – 49-81) and 60% of the patients were females. Most patients (80%) had an ECOG score of 1. Twelve patients (60%) had distant metastases, while the others had locally advanced disease. Median compliance with TTFields was 78% (14 hours/day), with median duration of 5 months. Fourteen patients (70%) had serious (grade 3-5) AEs during the study period. Six patients (30%) had hematological, 45% gastrointestinal and 15% pulmonary AEs. Ten patients (50%) had treatment-related skin toxicity, of which only 2 were grade 3, both resolved with appropriate treatment. No TTFields-related serious AEs were reported. The median PFS was 8.3 months (95% CI 4.3, 10.3). PFS rate at 6 months was 56%. Of the evaluable tumors, 30% had partial response and another 30% stable disease. The median OS was 14.9 months and 1-year survival rate was 55%. Conclusions: TTFields concomitant to gemcitabine are tolerable and safe for advanced pancreatic cancer patients. The efficacy results are promising and support further research in this indication. An extension of the PANOVA protocol, including 20 additional patients who receive gemcitabine, nab-paclitaxel and TTFields is ongoing. Clinical trial information: NCT01971281.

Inactivation of MLH1 due to promoter hypermethylation strongly suggests a sporadic origin, providing exclusion criteria for Lynch syndrome. The aim of this study is to compare the utility of methylation analysis of MLH1 and BRAF V600E mutations for the selection of patients with MLH1 negative colorectal cancer for genetic testing. MLH1 methylation status was evaluated by MethyLight and methylation-specific MLPA (MS-MLPA) in tumor DNA from 73 colorectal cancer patients with loss of MLH1 protein expression. These tumors were analyzed for BRAF V600E mutations, and genetic testing for germline MLH1 mutations was performed in all corresponding patients. Ten patients had germline mutations in MLH1 and none of their tumors showed significant MLH1 methylation or BRAF V600E mutation. MLH1 genetic testing excluded patients by MethyLight in 47 patients (64%), by MS-MLPA in 49 (67%), and BRAF V600E mutation in only 25 patients (34%) (χ2P = 0.00001). Specificity was 75% for MethyLight, 78% for MS-MLPA and 40% for BRAF V600E mutation. The use of MethyLight or MS-MLPA instead of BRAF mutation resulted in a cost reduction of 41% and 45%, respectively, per every MLH1 mutation detected. Taken together, methylation analysis of MLH1 shows better performance characteristics than BRAF V600E mutation in the selection of patients for genetic testing of MLH1, especially when using MS-MLPA. Supported in part by a grant from the Fundación de la CV para la investigación en el Hospital General Universitario de Alicante (Grupos consolidados 2008). L.P.-C. is the recipient of a grant from the Instituto de Salud Carlos III (FI07/00303). N.A. is supported by Schering-Plough.

Purpose: Immunohistochemistry for mismatch repair proteins has shown utility in the identification of Lynch syndrome, but majority of tumors with loss of MLH1 expression are due to sporadic hypermethylation of the MLH1 promoter. These tumors can also show epigenetic silencing of other genes, such as p16. The aim of our study is to evaluate the utility of p16 immunohistochemistry in the prediction of MLH1germline mutations. Experimental Design: p16 immunohistochemistry was appropriately evaluated in 79 colorectal cancers with loss ofMLH1expression.Methylation of MLH1 and p16 were quantitatively studied using real-time PCR assay Methylight. BRAF V600E mutation in tumor tissue was also investigated. Genetic testing for germline mutation of MLH1was made on 52 patients. Results: Loss of p16 expression was seen in 21 of 79 samples (26.6%). There was found statistically significant association between p16 expression and p16 methylation (P < 0.001), MLH1 methylation (P < 0.001), and BRAF mutation (P < 0.005). All tumors with loss of p16 expression showed hypermethylation of p16 (21of 21), 95.2% (20 of 21) showed MLH1 methylation, and 71.4% (15 of 21) were mutated for BRAF V600E. Mutational analysis showed pathogenic germline mutations in 8 of the patients, harboring 10 tumors. All 10 of these tumors showed normal staining of p16 in the immunochemical analysis. Conclusions: p16 immunohistochemistry is a good surrogate marker for p16 and MLH1 epigenetic silencing due to hypermethylation, and is useful as screening tool in the selection of patients for genetic testing in Lynch syndrome. Generalitat Valenciana Conselleria de Sanitat (AP 021/07), Fundación de la CV para la Investigación en el Hospital General Universitario de Alicante (2008). Beca predoctoral Instituto de Salud Carlos III (FI07/00303, Lucía Perez-Carbonell).

9 páginas, 1 figura, 4 tablas.-- et al., [Purpose]: It is not clear that the published estimates of the breast and ovarian cancer penetrances of mutations in BRCA1 and BRCA2 can be used in genetic counseling in countries such as Spain, where the incidence of breast cancer in the general population is considerably lower, the prevalence of BRCA2 mutations seems to be higher, and a distinct spectrum of recurrent mutations exists for both genes. We aimed to estimate these penetrances for women attending genetic counseling units in Spain. [Experimental Design]: We collected phenotype and genotype data on 155 BRCA1 and 164 BRCA2 mutation carrier families from 12 centers across the country. Average age-specific cumulative risks of breast cancer and ovarian cancer were estimated using a modified segregation analysis method. [Results]: The estimated average cumulative risk of breast cancer to age 70 years was estimated to be 52% [95% confidence interval (95% CI), 26-69%] for BRCA1 mutation carriers and 47% (95% CI, 29-60%) for BRCA2 mutation carriers. The corresponding estimates for ovarian cancer were 22% (95% CI, 0-40%) and 18% (95% CI, 0-35%), respectively. There was some evidence (two-sided P = 0.09) that 330A>G (R71G) in BRCA1 may have lower breast cancer penetrance. [Conclusions]: These results are consistent with those from a recent meta-analysis of practically all previous penetrance studies, suggesting that women with BRCA1 and BRCA2 mutations attending genetic counseling services in Spain have similar risks of breast and ovarian cancer to those published for other Caucasian populations. Carriers should be fully informed of their mutation- and age-specific risks to make appropriate decisions regarding prophylactic interventions such as oophorectomy.

242 Background: The prognosis of patients diagnosed with Pancreatic Cancer (PC) is dismal with a 5-year survival rate of around 5%. Familial Pancreatic Cancer (FPC) is an autosomal dominant rare syndrome defined as families with two or more first-degree relatives with pancreatic cancer that do not fulfill the criteria of any other inherited tumor syndrome. The Spanish familial pancreatic cancer registry, Pan-Gen-FAM was established in 2009 in order to identify and manage families at high risk of developing PC. Methods: Information on the family history of cancer is collected in order to determine the phenotype of individual families and patients are offered genetic testing of known FPC associated genes pertinent to their familial syndrome. An individualized clinical screening program is devised for the early detection of a pancreatic tumor consisting of periodic monitoring by imaging techniques (EUS and MRI) and the evaluation of minimally-invasive tumor biomarkers approaches including Circulating Tumor Cells (CTC) and circulating free DNA in blood. Results: To date the registry includes 125 individuals representing some 41 families. Of 17 families tested so far BRCA2 germline mutations were detected in 5 families and a BRCA1 mutation in 1 family. 43 individuals underwent clinical screening. More pancreatic abnormalities were found by EUS (41%) than by MRI (31%). The most frequent abnormal findings were parenchymal changes associated with chronic pancreatitis. Eight cystic lesions were identified. One lesion was identified as a well differentiated neuroendocrine tumor and another was shown to have benign cytology and whilst carcinoma was found in a third patient, this patient underwent a subsequent partial pancreatectomy. The remaining patients with cystic lesions are undergoing close clinical observation. Seventy three patients underwent CTC determinations and all patients tested negative. Conclusions: Periodic screening of high risk individuals by EUS and RMI can detect small tumors and premalignant lesions. We hypothesize that these persons will have a higher probability of long-term survival than sporadic cases.

201 Background: Advanced gastric cancer (AGC) is a highly lethal disease worldwide with a median overall survival (OS) of less than 12 months. Ramucirumab was the first targeted agent to show survival benefit in the second-line (2L) treatment of AGD. From the REGARD trial, women exhibited a trend towards a worse survival when compared to men. Methods: We retrospectively analyzed the outcome of a Spanish cohort of 55 AGC patients who had received at least two consecutive chemotherapy lines of treatment at our institution. We preselected some clinical parameters (age, gender, Lauren subtype, Her-2 status, grade of differentiation, primary tumor localization and resection, TNM stage, neoadjuvant/adjuvant administration, mono/polychemotherapy in 2L, irinotecan- or taxane-based 2L, Trastuzumab treatment, and administration of third-lines [3L]) to be tested as potential prognostic factors. We focused our analysis on the gender effect in OS after 2L treatment in AGC. The SPSS software package was used for all statistical analyses (Mann-Whitney U test, Chi Square, Kaplan-Meier, Cox regression). Results: Women, diffuse subtype, poorly differentiated tumors, sub-cardia location, single-agent 2L and 3L administration were the clinical characteristics with a significant shorter OS by the univariable analysis. Median OS of women (5 months, 95%CI 3.9-6.0) was significant lower than that in men (13 months, 95%CI 7.9-18.0) by both Log-Rank and Breslow tests (p 0.002). We found no significant differences in age, progression-free survival or time-on-treatment between men and women. Female gender was significantly associated with more taxane-based 2L and with less 3L treatment administration. When 3L treatment administration was excluded in the multivariable analysis, female gender was the only independent prognostic factor of all the clinical parameters included in our review (HR 0.33, 95%CI 0.15-0.75, p 0.008). Conclusions: Our retrospective study found that gender, Lauren subtype, grade of differentiation, primary tumor location, polychemotherapy and 3L administration may be potential prognostic factors in AGC after 2L treatment. Female gender was an independent prognostic factor in our Spanish cohort.

e15019 Background: Overall survival (OS) is considered the traditional endpoint for assessing the efficacy of first-line treatments in clinical trials with advanced gastric cancer. Progression free...

3560 Background: Downsizing with chemotherapy (CT) unresectable liver-limited disease (LLD) can permit resection and prolonged survival in patients (pts) with colorectal cancer (CRC). We assessed t...

4122 Background: G and E have shown a survival benefit in the 1st-line setting in mPC. The aim of this study was to assess whether combining C with G+E was safe and effective as compared with G-E i...

1531 Background: Gender differences in lung cancer (LC) have been reported, but with many unresolved issues yet. Tobacco causes the majority of women lung cancer (WLC), although the rate of never-smoking WLC is higher than in men. Several factors may play etiologic roles, and an in-depth understanding is needed. Methods: WORLD07 is a Spanish prospective, multicenter, epidemiologic female-specific LC database sponsored by ICAPEM, a professional association committed with WLC research. In order to improve the knowledge on never-smoking WLC, information has been extracted from WORLD07 database. Results: From October/2007 to October/2011, 1371 newly diagnosed WLC were included in an e-database from 32 centers, 539 (39.3%) never-smoking. Patient (p) characteristics: median age 71.1 years(y); median age of menarche 13y.; motherhood 91.2% (median 2.3 children, median age at first child 26.4y); oral contraceptive use 11.9%; postmenopausal 88.9% (median age of menopause 49y); HRT 5.2%; second-hand smokers 40% (work-exposure 17.1%, home-exposure 88.8%); obesity 16.3%; familiar history of cancer 39.9% (LC 29.8%); previous history of cancer 13% (breast/lung/cervix: 41.4/5.7/2.9%); current LC histology(%): adenocarcinoma/SqCC/LCC/SCLC: 83.4/6.2/5.5/3.9; EGFR mut+ (268 p analyzed): 55.5% (exon 19/20/21(%): 61.1/7.4/36.9); TNM NSCLC I/II/III/IV(%): 14/3.3/19.8/60.3. Treatment: EGFR-TKI in p EGFR mut+, stage IV(1st-/2nd-line)(%): 51.7/15.4; stage IV NSCLC (1st-line/2nd-line): platinum-based CT 42.5%, EGFR-TKI 33.5%, combinations with bevacizumab 2.9% / EGFR-TKI 15.8%. Overall survival: median 27 months(m), 1/2-y(%) 74.8/55.2; stage IV NSCLC: median 20.5m, 1/2-y(%) 67/46; EGFR-mut+ p: median 27.3m, 1/2-y(%) 75/54.3. Conclusions: According to our e-database, WLC showed high rates of never-smokers (39.3%), and of relatives diagnosed with malignant tumours (39.9%, ≅1/3 LC). Adenocarcinoma was the most frequent histology (76.1%), and more than half of the cases analyzed harboured EGFR mutations. Although 40% were second-hand smokers, further investigations are warranted. Survival outcomes remain satisfactory, as expected from this selected subgroup of p. Additional epidemiologic and treatment data will be presented.

e22077 Background: Different predictive models for Lynch syndrome have recently been developed and their comparative performance in a clinic-based cohort has not been assessed. We aimed to analyze the accuracy of the MMRpro, Barnetson, and PREMM1,2 models to predict MLH1/MSH2 mutation carrier status in 564 unrelated probands with clinical suspicion of hereditary colorectal cancer and compare it with Wijnen model and clinical criteria. Methods: Overall, 538 individuals (95%) underwent mismatch repair (MMR) deficiency screening before germline genetic testing (sequencing with or without large rearrangement analysis) and 26 (5%) performed direct genetic testing. Prediction scores for all individuals were calculated by each model. Sensitivity, specificity, positive predictive value (PPV), and the areas under the receiver operating characteristics curves (AUC) for all models were calculated and compared with the Revised Bethesda Guidelines (RBG). Results: 114 individuals (20%) were mutation carriers (63 MLH1, 51 MSH2). The AUC was 0.95 (95% CI: 0.93–0.97) for MMRpro, 0.87 (95% CI 0.83–0.91) for the Barnetson model, 0.87 (95% CI 0.83–0.91) for PREMM1,2, and 0.75 (95% CI 0.69–0.80) for the Wijnen model (p1,2. Sensitivity and specificity of RBG were 86% and 14%, respectively. Calibration was 0.92, 1.05, 0.50, and 1.25 for PREMM1,2,Barnetson, Wijnen, and MMRpro, respectively. Conclusions: In a population of individuals at risk of Lynch syndrome, the MMRpro model has the largest AUC, although the Barnetson and PREMM1,2 model also show adequate discrimination. Any of the models perform better than the RBG and provide quantitative risk estimation of finding a MLH1/MSH2 mutation useful in genetic counselling. No significant financial relationships to disclose.

8095 Background: Pemetrexed is approved as second-line therapy in advanced NSCLC. In this study, we evaluate the efficacy and safety of pemetrexed in patients with Advanced NSCLC previously treated and we also perform a pharmacogenomic analysis to determine if biologic characteristics could be related to any clinical benefit. Methods: 195 patients with NSCLC stage IIIB (malignant pleural effusion) or IV were selected to receive pemetrexed (500 mg/m2, IV infusion) supplemented with vitamin B12, folic acid and dexamethasone prophylaxis. DNA was extracted from peripheral lymphocytes and Taqman assays for allelic discrimination were used for the following single- nucleotide polymorphism (SNP) typing: Methylene-Tetra-Hydro-Folate-Reductase (MTHFR), methionine synthase (MTR), thymidylate synthase (TS) and reduced folate carrier 1 (RFC1). Results: Median age 63 years (37–82); males: 88%; never-smokers: 9%; performance status (PS) 0: 23%; adenocarcinoma: 34%, squamous: 45%; stage IIIB: 20%, IV: 80%. The responses...

15111 Background: Molecular studies have demonstrated a direct relationship between human papillomavirus (HPV) infection and invasive cervical cancer, as well as cancer of the anal region. The role...

815 Background: The addition of oxaliplatin to fluorouracil and leucovorin as adjuvant therapy for patients with stage II and III colon cancer (CC) has been analyzed in two large, randomized trials, MOSAIC and C-07 trials. The updated results of these studies showed that the addition of oxaliplatin enhances overall survival by approximately 5% in patients with stage III disease but has no effect in patients with stage II disease. Methods: We retrospectively included patients with stage II and III CC that were operated between 2009 and 2014 in the Ramón y Cajal University Hospital from Madrid. We perform a multivariable Cox model analysis to estimate the benefit of the chemotherapy stratifying by oxaliplatin in each stage. The model was further adjusted by including the following confounders: ECOG-PS, number of removed nodes, perforation, obstruction, grade and age. Stata 13.1 was used to analyze the data. Results: 564 patients were identified (281 stage II and 283 stage III). 305 did not receive any chemotherapy, 61 received monotherapy with fluoropyrimidines (FP) and 198, FP and oxaliplatin. The median follow-up in the entire cohort was 49 months. Globally, adjuvant chemotherapy (either with FP alone or with the combination with oxaliplatin) showed no benefit in DFS (HR of 1.18 and 0.98, respectively). The benefit in OS was significant either for FP alone (HR: 0.46, p: 0.029) and for the combination treatment (HR: 0.41, p: 0.001). Patients with stage II treated with FP in monotherapy showed no benefit, neither in DFS nor OS (HR for DFS: 2.2, p: 0.1; HR for OS: 0.5, p: 0.22). The benefit was neither seen with the addition of oxaliplatin (HR for DFS: 2, p: 0.11; HR for OS: 0.85, p: 0.7). Stage III patients treated with FP presented a HR for DFS of 0.76 (p: 0.5) and a HR for OS of 0.42 (p: 0.087). The HR for DFS with oxaliplatin was 0.53 (p: 0.07). A significant improvement in OS was observed, with a HR of 0.22 (p < 0.001). Conclusions: The addition of oxaliplatin in the adjuvant treatment of stage III patients showed a trend towards improvement in DFS and a significant benefit in OS compared to not receiving chemotherapy. On the contrary, patients with stage II did not benefit from this treatment, neither in DFS nor OS.

842 Background: The standard adjuvant treatment for patients with stage III colon cancer is 6 months with fluoropyrimidines and oxaliplatin. The Intergroup Trial INT-0035 was the first large-scale study to demonstrate a significant reduction in the risk of death with adjuvant FU plus levamisole in patients with stage III colon cancer. In the MOSAIC study, the addition of oxaliplatin to fluoropyrimidines in patients with resected stage II to III colon cancer showed OS and DFS benefit of oxaliplatin. However, no significant benefit was observed in either DFS or OS in patients with stage II disease, therefore the benefit of adjuvant chemotherapy is still controversial in those patients. Methods: We retrospectively included patients with stages II and III colon cancer that were operated between 2009 and 2014 in the University Ramón y Cajal Hospital from Madrid. We calculate DFS and OS at 48 months and we perform a multivariable Cox model analysis to estimate the benefit of the chemotherapy in each stage. The model was further adjusted by including the following confounders: ECOG-PS, number or removed nodes ( < 12 or ≥ 12), grade and age. A covariate was considered a confounder factor if the difference between the adjusted and unadjusted coefficient of chemotherapy varied > 10%. Stata 13.1 was used to analyze the data. Results: 564 patients were identified (281 stage II and 283 stage III). 259 received chemotherapy and 305 did not. The median follow-up in the entire cohort was 49 months. The median DFS and OS were not reached at the moment of the analysis. DFS and OS at 48 months were both 78.5%. Globally, chemotherapy did not improve DFS (HR 1.05, p: 0.83) but OS was significantly better (HR 0.47, p: 0.001). By stage, chemotherapy did not improve DFS in stage II (HR: 1.6, p: 0.2) nor OS (HR 0.76, p: 0.43). In stage III, chemotherapy showed a trend to improve DSF (HR: 0.61, p: 0.075) and did improve OS (HR: 0.31, p < 0.0001). Conclusions: The multivariable analysis showed a chemotherapy benefit in patients with stage III colon cancer, with a 39% reduction in the risk of recurrence and a 69% in the risk of death; however, in stage II patients these benefits were not found either in DFS or OS.