ABSTRACT
AIMS: Lung metastasectomy and, more recently, stereotactic body radiotherapy (SBRT), are frequently proposed to stage IV oligometastatic colorectal cancer (CRC) patients. In the absence of a randomised comparison between the two treatments, we aimed to retrospectively explore the effect on overall survival and progression-free survival (PFS) in a comparative cohort study. MATERIALS AND METHODS: We included patients who consecutively underwent surgery (n = 142) or SBRT (n = 28) as first local therapy at the time of lung progression, between 2005 and 2012. Both overall survival and PFS functions according to treatment were calculated using the Kaplan-Meier method and compared using the Log-rank test. The effect of treatment on overall survival and PFS was estimated by Cox models using different adjustment methods. RESULTS: Patients receiving SBRT were older and were treated more recently, whereas the two cohorts were similar for most baseline prognostic factors. Overall survival at 1 and 2 years was 0.89 and 0.77 for SBRT and 0.96 and 0.82 for surgery (P = 0.134), respectively. Multivariable analyses did not highlight a clear treatment effect on overall survival (adjusted hazard ratioSBRT versus surgery = 1.71; 95% confidence interval 0.82-3.54; P = 0.149) and even smaller differences using the inverse probability treatment weighting method (hazard ratioSBRT versus surgery = 1.28, 95% confidence interval 0.58-2.82; P = 0.547). The results of PFS were unreliable because different follow-up protocols were applied in the two cohorts. CONCLUSION: With limitations consisting in the retrospective observational design and different sample sizes, the results of this explorative analysis indicate that overall survival probability after SBRT is similar to surgery for the first 2 years from treatment. This finding supports the need for high-quality trials comparing different treatment modalities for lung oligometastases from CRC.
Subject(s)
Colorectal Neoplasms/pathology , Colorectal Neoplasms/radiotherapy , Colorectal Neoplasms/surgery , Lung Neoplasms/radiotherapy , Lung Neoplasms/secondary , Lung Neoplasms/surgery , Aged , Cohort Studies , Disease Progression , Disease-Free Survival , Female , Humans , Male , Middle Aged , Pneumonectomy , Proportional Hazards Models , Radiosurgery/methods , Retrospective Studies , Survival AnalysisABSTRACT
BACKGROUND: Early tumor shrinkage (ETS) and depth of response (DoR) predict overall survival (OS) in first-line trials of chemotherapy ± anti-EGFR monoclonal antibodies in metastatic colorectal cancer (mCRC). These associations and the predictive accuracy of response measurements for survival parameters were investigated in the phase III TRIBE trial of FOLFOXIRI plus bevacizumab (bev) versus FOLFIRI plus bev. PATIENTS AND METHODS: A landmark approach was adopted to define the assessable population. The distribution of RECIST response rate, ETS and DoR was compared in the two arms. Associations between response measurements and progression-free survival (PFS), post-progression survival (PPS) and OS were tested by univariate and multivariate Cox models. Prediction performance of each factor was estimated by C-index. RESULTS: A significantly higher percentage of patients in the FOLFOXIRI plus bev arm achieved ETS ≥20%, when compared with the control arm (62.7% versus 51.9%, P = 0.025). Also the DoR was significantly higher in the triplet plus bev arm (43.4% versus 37.8%, P = 0.003). Both ETS and DoR were associated with PFS, PPS and OS at the univariate analyses and in the multivariate models stratified for other prognostic variables. Both ETS and DoR were able to predict survival as accurately as RECIST response. CONCLUSION: FOLFOXIRI plus bev improves ETS and DoR when compared with FOLFIRI plus bev. Achieving rapid and deep tumor shrinkage consistently delays tumor progression and prolongs survival in patients treated with first-line chemotherapy plus bev. ETS is a promising and valuable end point for clinical trials' design deserving further investigation.