Search

e12539 Background: The Notch signaling pathway plays an important role in cell-differentiation, survival, proliferation, stem-cell renewal, and in determining cell fate during development and morphogenesis. The dysregulation of the Notch pathways contributes to carcinogenesis, cancer stem cell renewal, angiogenesis, and chemo-resistance. Elevated levels of Notch receptors and ligands have been associated with cancer-progression and poor survival. A less explored function of Notch pathways in cancer is their role in leukocyte homing and activation. Understanding their role and relationship with immune infiltrates is an area of interest in cancer research. Methods: The expression of four different Notch genes (Notch1, Notch2, Notch3 and Notch4) was explored in relation with A luminal breast cancer patient outcome using transcriptomic data (Affymetrix dataset, exploratory cohort) and the METABRIC study (validation cohort). The TIMER online tool was used to explore the association of the identified notch and immune infiltration, and the TCGA and METABRIC studies to analyze the correlation between notch1 - 4 expression and genomic signatures of immune activation. Results: We identified 2 individual genes called Notch1 and Notch2, which predict favorable prognosis in luminal A breast cancer. Their expression positively correlated with the presence of immune infiltrates within the tumor (dendritic cells, CD4+ T cells, neutrophils, CD8+ T cells and B cells), with markers of T cell activation and antigen presentation, and with gene signatures of immune surveillance (cytotoxic T lymphocyte activation and IFN gamma signature). By contrast Notch3 and Notch4 which predicted for detrimental outcome were not associated with any of these parameters. Conclusions: Our analysis identifies a Notch signature composed of 2 genes with potential to recognize immune infiltrated and activated A luminal phenotype breast cancers with favorable prognosis.

564 Background: The study of sentinel lymph nodes (SLN) assessed by One Step Nucleic Acid Amplification (OSNA, Sysmex, Kobe, Japan) creates a new variable, Total Tumor Load (TTL). This variable is defined as the total number of CK19 mRNA copies in all positive SLN (copies/microL). The latest edition of the Spanish Oncological Gynecology Society (SEGO) Guideline (2017) proposes a complete axillary lymph node dissection (ALND) when TTL is 15,000 copies or more in early breast cancer. In our center we are using OSNA to ascertain if there is axillary node involvement but the decision to proceed to ALND is based on Z0011 criteria. We want to determine if there is a correlation between clinical outcomes and TTL values, between TTL and pathological variables and if TTL is a useful tool to decide when to complete an ALND. Methods: Clinicopathological and follow up data were obtained from all patients with invasive breast cancer and SLN assessed by OSNA between 2011 and 2017 at our center. Results: A total of 321 patients underwent SNB assessed by OSNA with an average follow-up of 56 months. 320 were female and 1 male. Age range 27-89 years (mean 58.9). 85 % were ductal, 10 % lobular and 5 % other. 53.5% were luminal A, 28.66% luminal B, 7.78%, triple negative, 4.3%, Her2 positive and 4.3%. luminal B-Her2 positive.TTL was equal to 0 in 183 cases and greater than zero in 138 cases.71 cases showed a TTL higher than 15,000 copies. Only 21 cases met Z0011 criteria and had ALND. As of now, 3 patients have had locoregional relapse and 8 metastatic disease. 12 have died, only two from metastatic breast cancer. Conclusions: Using Z0011 criteria, we have adequate clinical outcomes with a low rate of ALND; If we had based the axillary management on TTL values we would have multiplied the number of ALND by a factor of 3.3 (from 21 to 71); We have observed a tendency to higher TTL in luminal phenotypes and to lower TTL in HER2 positive and triple negative subtypes; Work is in progress to increase our sample size.

e20648 Background: The molecular study of non-small cell lung cancer (NSCLC) allows for targeted treatment. EGFR and ALK alterations have been observed in 14% and 5% of Caucasian patients (p) respectively. These alterations are mainly detected in women with adenocarcinoma (ADC) and non-smokers. In a non-selected (all comers) NSCLC population from Ávila (Spain) we found a different profile. Methods: We collected data from 185 consecutive p the Pathology Department database of Hospital Nuestra Señora de Sonsoles from 2012 to 2017. All p were tested for EGFR and ALK. Analyses were done with Cobas or Therascreen tests. Results: EGFR mutations were found in 11 p out of 185 (6.7%). Histology was ADC in 8 p (72.7%) and squamous cell carcinoma (SC) in 3 p (27.3%). Male (m) & smokers: 4/11 (36,4%). The type of mutation was: Exon 20: 3 p (27.3%). 2 m and 1 female (f). 2 smokers and 1 former smoker. 2 ADC and 1 SC; Del 19: 4 p (36.4%). 1 m and 3 f. 1 non-smoker, 2 former smokers and 1 smoker. 3 ADC and 1 SC; Exon 21: 4 p (36,4%). 2 m and 2 f. 3 non-smokers and 1 smoker. 3 ADC and 1 SC. All p. received treatment with 1st generation anti-EGFR drugs for a median 6.4 months (m). Progression Free Survival was 6.39 m (exon 20: 5.1 m; Del 19: 6.5 m; exon 21: 7.2 m). Overall survival was 18.0 m (exon 20: 26.1 m; Del 19: 18.2 m; exon 21: 11.8 m). ALK translocation was found in 5 p out of 158 (3.2%). Histology was ADC in 3 p (60%) (all f, 2 non-smokers and 1 former smoker) and SC in 2 p (40%) (2 smoker m.) Conclusions: Although the small sample size prevents definitive conclusions, EGFR mutations in exon 20 are more frequent in smokers and mutations in exon 21 in non-smoking patients with ADC. ALK translocation was observed independent of sex, tobacco or histologic type. For EGFR and ALK alterations, screening based on sex or pathology would have missed findings with therapeutic implications. We conclude it is important to extend the analysis of molecular alterations to all patients with NSCLC since more of them could benefit from targeted drugs.[Table: see text]

e12574Background: The study of sentinel lymph nodes (SLN) assessed by One Step Nucleic Acid Amplification (OSNA, Sysmex, Kobe, Japan) creates a new variable, Total Tumor Load (TTL). This variable i...

e12062 Background: Prosigna is a standardized test based on the PAM50 gene signature which provides information on intrinsic subtype and risk of recurrence (ROR) score predicting 10y recurrence probability (NanoString Technologies, Inc., Seattle, WA). We evaluated Prosigna’s impact on adjuvant treatment decision beyond immunohistochemistry (IHC) testing. Methods: 125 pre- and postmenopausal EBC patients with ER+, HER2-, pT1-T2 pN0/pN1mi were enrolled (12 centers; 8/2015-11/2016). FFPE specimens were centrally analyzed using Prosigna to classify patients according to the intrinsic tumor subtype and ROR score. The primary endpoint was the impact of the Prosigna test on adjuvant treatment decision. Results: 64% of tumors were classified by PAM50 as Lum A, 35% as Lum B, 0% Basal, 1% HER2-E. The intrinsic subtype concordance between immunohistochemistry (IHC) and Prosigna (n = 119) was 61.3%. In Lum B tumors by IHC the discordance rate was the highest (48% were reclassified as Lum A by Prosigna) as shown in the Table. ROR risk groups were as follows: ROR low (54; 43%), ROR interm. (38; 30%), and ROR high (33; 27%). Prosigna results led to a treatment recommendation change in 49 (39%) patients. In the 76 (61%) cases with the initial recommendation of Hormonal Therapy (HT) alone the final decision changed to chemotherapy (CT)+HT (CHT) in 24 (32%) patients. In the 49 (39%) cases in which the initial recommendation was CHT the final decision changed to HT in 25 (51%) patients. Among the 38 patients with Prosigna intermediate risk, 21 (55%) were allocated to HT. Conclusions: In this prospective decision impact study, Prosigna results led to a 39% change in adjuvant therapy indication. Patients with initial indication of CHT were changed to HT alone in > 50% of cases. Thus, Prosigna results influenced the treatment decisions and reinforced its clinical utility in real-world settings. The intrinsic subtype classification based on IHC didn’t show to be an adequate surrogate for the genomic subtypes as determined by Prosigna. [Table: see text]

e21521Background: In our institution, median age of patients with lung cancer is 71. We decide to study how are treated patients over 80 years. Methods: We performed a retrospective observational s...

638 Background: Head-to-head comparison of a steroidal aromatase inhibitor (AI) (E) vs. a non-steroidal AI (A) with crossover at the time of progression to evaluate the sequence of administration. Principal and secondary objectives were overall response rate (ORR), clinical benefit (CB) and time to progression (TTP). Methods: Postmenopausal pt with measurable ABC and positive HR, were stratified [previous tamoxifen (tam) (yes vs. no), previous chemotherapy (ct) (yes vs. no) and stage (IIIb vs. IV)], and randomized to E (25mg/day), or A (1mg/day) until progression. As per Simon’s test, with a p1 of 25% ORR and a p0 of 10% ORR, α and β errors 0.05 and 10%, 50 pt per arm were needed. Results: 103 (51 vs. 52) pt were recruited between 2002 and 2004. Main pt characteristics: median age 70 and 73 years, 82 vs. 88% stage IV, 12 vs. 17% previous ct, 51 vs. 50% previous tam, median size of lesions of 5 vs. 4 cm, mean number of lesions of 1.4 in both arms. Most frequent site of lesions were breast (33 vs. 35%), lymph nodes (22% vs. 25%) and connective tissue (27% vs. 23%). Efficacy: For initial treatment ORR was 37% (CI95% 24–51) in arm E vs. 52% (CI95% 38–66) in arm A (p= 0.151). 47% of pt in E vs. 60% in A presented CB (p=0.202). Mean TTP for arm E was 8.1 months (m) (1.1–15.2) and for arm A was 12.1 m (6.6–17.6) (p=0.4716). 11 pt (arm E) and 12 (arm A) crossed-over after progression. Following crossover mean TTP for E was 4.4 m (3.7–5.1) and for second line A was 1.9 m (0.1–3.7) (p=0.0172). Conclusions: E and A seem to be equivalent in terms of efficacy as first treatment for postmenopausal pt with ABC and positive HR. We suggest a possible advantage of the steroidal AI, exemestane, following disease progression on the non-steroid AI anastrozole. [Table: see text]