Recently, phase I studies with novel antibody drug conjugates targeting HER2 suggested benefit in HER2-low patients -defined as immunohistochemistry(IHC) +1 or +2 FISH/ISH non-amplified, with advanced breast cancer(BC). Data on the prognostic value of HER2-low in early stage disease is scarce. The purpose of this study was to evaluate the impact of HER2-low status on response to neoadjuvant chemotherapy(NACT) and survival outcomes in early stage HER2negative BC. MethodsRecords from all BC patients treated with NACT from January 2007 to December 2018 in a single cancer center were retrospectively reviewed. Primary objective was to compare differences between pathologic complete response(pCR) and relapse free survival(RFS) in luminal HER2-low/HER2-0 and triple negative(TNBC) HER2-low/HER2-0. Results855 non-HER2-positive patients were identified. Median follow-up was 59 months. 542 had luminal BC (63.4%) and 313 TNBC (36.6%). 285 (33.3%) were HER2-low. Among luminal tumors, 145 had HER2 IHC+1 (26.8%) and 91 IHC+2/ISH non-amplified (16.8%). In TNBC, only 36 had HER2 IHC+1 (11.5%) and 13 IHC+2/ISH non-amplified (4.2%). Among luminal/HER2-low and luminal/HER2-0 population, there was a high proportion of clinical T3/4 (61.5% vs 69.2%, p=0.053), node positive (74.2% vs 66.3%, p=0.27) and stage III tumors (63.1% vs 65%, p=0.51). The same was true TNBC/HER-low as compared to TNBC/HER2-0, despite a non-statistically significant higher cT4 among TNBC/HER-low (32.7% vs. 19.3%, p=0.17). pCR was 13% in luminal/HER2-low versus 9.5% in luminal/HER2-0 (p=0.27), and 51% in TNBC/HER2-low versus 47% in TNBC/HER2-0 (p=0.64). 5y RFS was 72.1% in luminal/HER2-low and 71.7% in luminal/HER2-0 (p=0.47), and 75.6% in TNBC/HER2-low versus 70.8% in TNBC/HER2-0 (p=0.23). HER2-low status was not associated with RFS in multivariate analysis (HR 0.83, 95%CI 0.6-1.11, p=0.21). ConclusionOur data does not support HER2-low as a biologically distinct BC subtype, with no predictive effect on pCR after NACT nor prognostic value on survival outcomes.

presented an event of interest for DDFS analyses. Five-year DDFS was 91,6%. Predictors for DDFS in univariate analysis were cT stage (c2¼45.408, p<0.001), AJCC Anatomic Stage Groups (c2¼ 110,784, p<0.001), and AJCC Clinical Prognostic Stage Groups (c2¼ 22,868, p¼0.001). In the multivariate analyse only AJCC anatomic stage groups retained significative results for DDFS (p<0,001).Conclusions: In our study, after pCR in breast cancer patients treated with NACT, only AJCC anatomic stage groups was significant for DDFS in the multivariate analyse. In this special population, these results emphasize the importance of meticulous staging at diagnosis and suggest the possibility of personalised follow-up and modulated adjuvant treatment after surgery based on AJCC anatomic stage groups.Legal entity responsible for the study: The authors.

Purpose: Recently, phase I studies with novel antibody drug conjugates targeting HER2 suggested benefit in HER2-low patients – defined as immunohistochemistry(IHC) +1 or +2 FISH/ISH non-amplified, with advanced breast cancer(BC). Data on the prognostic value of HER2-low in early stage disease is scarce. The purpose of this study was to evaluate the impact of HER2-low status on response to neoadjuvant chemotherapy(NACT) and survival outcomes in early stage HER2- negative BC. Methods: Records from all BC patients treated with NACT from January 2007 to December 2018 in a single cancer center were retrospectively reviewed. Primary objective was to compare differences between pathologic complete response(pCR) and relapse free survival(RFS) in luminal HER2-low/HER2-0 and triple negative(TNBC) HER2-low/HER2-0. Results: 855 non-HER2-positive patients were identified. Median follow-up was 59 months. 542 had luminal BC (63.4%) and 313 TNBC (36.6%). 285 (33.3%) were HER2-low. Among luminal tumors, 145 had HER2 IHC+1 (26.8%) and 91 IHC+2/ISH non-amplified (16.8%). In TNBC, only 36 had HER2 IHC+1 (11.5%) and 13 IHC+2/ISH non-amplified (4.2%). Among luminal/HER2-low and luminal/HER2-0 population, there was a high proportion of clinical T3/4 (61.5% vs 69.2%, p=0.053), node positive (74.2% vs 66.3%, p=0.27) and stage III tumors (63.1% vs 65%, p=0.51). The same was true TNBC/HER-low as compared to TNBC/HER2-0, despite a non-statistically significant higher cT4 among TNBC/HER-low (32.7% vs. 19.3%, p=0.17). pCR was 13% in luminal/HER2-low versus 9.5% in luminal/HER2-0 (p=0.27), and 51% in TNBC/HER2-low versus 47% in TNBC/HER2-0 (p=0.64). 5y RFS was 72.1% in luminal/HER2-low and 71.7% in luminal/HER2-0 (p=0.47), and 75.6% in TNBC/HER2-low versus 70.8% in TNBC/HER2-0 (p=0.23). HER2-low status was not associated with RFS in multivariate analysis (HR 0.83, 95%CI 0.6–1.11, p=0.21). Conclusion: Our data does not support HER2-low as a biologically distinct BC subtype, with no predictive effect on pCR after NACT nor prognostic value on survival outcomes.

644 Background: The introduction of tyrosine kinase inhibitors (TKI) and recently immunotherapy has brought major survival benefits for metastatic renal cancer (mRCC) patients (pts). In Brazil, there is no approved 2nd line treatment for mRCC in the Public Health System (PHS). Our center is unique, because it provides care for both PHS and Private System (PrS) population, enabling us to make the comparison of overall survival (OS) of these pts. Methods: We retrospectively reviewed medical records of all mRCC pts treated with 1st line TKI at our service from 2007 to 2018. Categorial variables were compared by Fisher’s exact test and continuous by Mann–Whitney. Survival was estimated by Kaplan-Maier method, prognostic factors adjusted by Cox regression model. Results: 171 pts were eligible, 37 (21.6%) PHS and 134 (78.4%) PrS pts. Between the two groups, there was no differences in age, gender, number and sites of metastasis (mets). PHS pts had worst ECOG (≥2, in 35.1 vs 13.5%, p .007), a trend towards more poor IMDC risk (IMDC favorable 16.2 vs 26.6%, intermediate 51.4 vs 57%, poor 32.4 vs 16.4%, p.09), had less nephrectomies (73 vs 92.5%, p.008) and more non clear cell histology (32.4 vs 12.7%, p.01). Median lines of therapy were 1 for PHS vs 2 for PrS pts (p.03). Sunitinib was the 1st line agent for 91.9 vs 67.2%, and pazopanib 8.1 vs 29.9%, of the PHS and PrS pts, respectively. Median time from diagnosis of mets to treatment start was 2.29 vs 1.79 m (p.59). Median OS was 16.5 vs 26.5 m (p.0002) and progression free survival, 8.4 vs 11 m (p.01), for the PHS vs PrS. On multivariate analysis, after adjusting for factors that were present before the beginning of treatment and were statistically significant for OS in the univariate model, PHS pts still had higher risk of death (HR 1.85, IC 95 1.2-2.9, p.01), probably due to having received fewer lines of therapy (≥2 lines of therapy vs 1, HR 0.51, IC95 0.4-0.7, p .001). Conclusions: Brazilian PHS pts had significant worse OS compared to PrS pts, in part due to less access to drugs. Access to cancer drugs is a challenge worldwide, and in Brazil effort has to be done to change this reality.

prolonged survival in chemo-refractory patients with mCRC. Here, we investigated the role of NLR as a predictive biomarker in the CAVE mCRC trial.Methods: 77 patients enrolled in the CAVE mCRC trial, treated with cetuximab rechallenge plus avelumab, were included in the current analysis. Baseline NLR was calculated and, in line with previous findings, a cut-off value of 3 was used. Plasma samples of 67 out of 77 patients were available for analysis of circulating tumor DNA (ctDNA) mutations of KRAS, NRAS, BRAF and EGFR-S492R. Correlation of NLR<3 vs NLR 3 with median overall survival (mOS), median progression-free survival (mPFS) in the intention to treat (ITT) and ctDNA RAS/BRAF/EGFR wild-type (WT) population was performed.Results: In the ITT population, mOS was 11.6 months [95% Confidence Interval (CI), 8.4-14.8 months] and mPFS was 3.6 months (95% CI, 3.2-4.1 months). Patients with a baseline NLR < 3 (38/77, 49%) had a statistically significant improvement in mOS, 17.3 months (95% CI, 14.2-20.3), compared with patients with NLR 3 (39/77, 51%) that exhibited mOS of 8.9 months (CI 95% 6.4-11.4), (HR 0.44, CI 95% 0.25-0.76, P¼ 0.004). mPFS was 3.9 months (CI 95% 2.9-4.9) in patients with NLR < 3, compared to 3.5 months (CI 95% 2.3-4.6) in patients with NLR 3 (HR 0.71, CI 95% 0.44-1.13, P¼ 0.15). In the ctDNA WT population, mOS was not reached in the NLR < 3 group (23/ 48, 48%), whereas a mOS of 8.9 months (CI 95% 6.1-11.7), (HR 0.24, CI 95% 0.10-0.58, P¼ 0.001) was reported in patients (25/48, 52%) with NLR 3. A trend towards increased mPFS was observed in NLR < 3 population [5.2 months (CI 95% 2.9-7.6) vs 3.6 months (CI 95% 2.8-4.3) (HR 0.67, CI 95% 0.36-1.23, P¼ 0.19)]. Conclusions:In the CAVE mCRC study, baseline neutrophil to lymphocyte ratio < 3 was strongly correlated with improved survival and may represent a useful biomarker to predict response to retreatment with cetuximab plus avelumab.