P-275 - The choice of irinotecan or oxaliplatin-based chemotherapy in metastatic colo-rectal cancer as first line treatment: Impact on survival (ERCC1 and TOPO1 adjusted analysis)

Abstract

Introduction: The choice between Irinotecan and Oxaliplatin-based chemotherapy as first-line treatment in metastatic colo-rectal patients is a difficult task. No specific criteria, guidelines or recommendation were defined. The oncologist should take into account some important demographic, clinical, biological and molecular variables: ERRC1, TOPO1 expression, all-RAS and/or B-RAF mutations, tumor location, MMR, CIMP status, etc. Methods: The primary endpoint of this retrospective and exploratory analysis was survival, defined as time-to-event variable, from primary diagnosis to death or date of last visit (for censored patients). The impact of chemotherapy (Irinotecan or Oxaliplatin-based) was analysed using multivariate Cox regression analysis. A crude (unadjusted) univariate analysis was performed as first-step. Covariates as age, gender, tumor location (colon vs rectal), type of molecular (targeted) treatment (Bevacizumab vs Cetuximab vs none), ERCC1 and TOPO1 expression (as defined by immunohistochemistry: the product (H-score) between the percentage of positive cells (0 to 100%) and intensity of expression (o to 3)). Hazard ratios were calculated (including 95% confidence interval) for unadjusted univariate analyses (Irinotecan vs Oxaliplatin-based chemotherapy) and for multivariate analyses having as covariate: a).ERCC1; b).TOPO1; c).ERCC1 + TOPO1; d).ERCC1 + TOPO1 + age + tumor location + gender + molecular treatment. No correction for multiplicity of tests was performed. Results: The cohort of interest was represented by 65 patients with metastatic colorectal cancer, treated between May 2011 - March 2017 in Saint-Constantin Hospital, Medical Oncology Department (Brasov, Romania). Gender distribution: 60% males and 40% women; tumor location: 77% colon and 23% rectal cancer. Chemotherapy regimens were Irinotecan (78%) and Oxaliplatin-based (22%). Median age was 60 years (range 37-81). Forty percent of patients were treated with Bevacizumab, 32% with Cetuximab and 28% by chemotherapy only. Median H-scores were 200 (ERCC1) and 190 (TOPO1). It seems that ERCC1 expression is generally bad (higher H-scores), compared with TOPO1 expression. At 71 months from the initial diagnosis, 53% of patients were still alive (95% CI, 21-85%). Hazard ratio for unadjusted univariate analysis (Irinotecan vs Oxaliplatin-based chemotherapy: HR = 1,81 (95% CI, 0,58-5,61), p-value=0,31) and for multivariate analysis having as covariate: a).TOPO1, HR = 1,99 (95% CI, 0,60-6,60), p-value=0,26; b).ERCC1, HR = 2,39 (95% CI, 0,73-7,80), p-value=0,15; c).ERCC1 + TOPO1, HR = 2,42 (95% CI, 0,70-8,33), p-value=0,16; d).ERCC1 + TOPO1 + age + tumor location + gender + molecular treatment, HR = 2,59 (95% CI, 0,67-9,97%), p-value=0,17 (as represented in Figure). ERCC1 had a powerful impact on survival (p-value=0,03), only. The risk of death for rectal cancer patients is 51% lower compared with colon cancer patients, no matter the type of chemotherapy and molecular treatment. Updated data will be presented at the end of June 2017. Conclusion: The probability of death is two and a half times lower in patients treated with Irinotecan-based regimens (FOLFIRI) compared to those treated with oxaliplatin (FOLFOX), as first-line therapy in metastatic setting. Hazard ratios were all (including 95% CI) above the “no effect” axis. Despite the low number of patients and a low statistical power, results clearly demonstrated the statistically significant predictive role of ERCC1 expression on survival in a multivariate Cox regression analysis. As showed, taking into account the ERCC and TOPO1 expression, FOLFIRI + molecular treatment should be the preferred regimen in first-line setting.