Perioperative Bevacizumab-based Triplet Chemotherapy in Patients With Potentially Resectable Colorectal Cancer Liver Metastases
2019
Abstract Background In colorectal cancer liver metastases (CRCLM), bevacizumab-based neoadjuvant strategies provide increased pathologic response. We aimed at assessing the activity of perioperative capecitabine, oxaliplatin, irinotecan, and bevacizumab (COI-B regimen) in patients with potentially resectable CRCLM, and investigating biomarkers for early prediction of pathologic response. Patients and Methods This was a single-center phase II study enrolling patients with liver-limited, borderline resectable disease and/or high-risk features. Patients received 5 preoperative and 4 postoperative cycles of biweekly COI-B (irinotecan 180 mg/m 2 and bevacizumab 5 mg/Kg on day 1, oxaliplatin 85 mg/m 2 on day 2, and capecitabine 1000 mg/m 2 twice a day on days 2 to 6). The primary endpoint was pathologic response rate in the intention-to-treat population. A Simon 2-stage design was adopted to detect an increase from 30% to 50% with a power of 90%. Dynamic imaging biomarkers (early tumor shrinkage [ETS], deepness of response, maximum standardized uptake volume [SUVmax]/regression index) and next generation sequencing data were explored as surrogates. Results From June 2013 to March 2017, 46 patients were enrolled. Pathologic response was achieved in 63% patients (endpoint met), and responders achieved significantly better survival outcomes with respect to non-responders. The most frequent grade 3/4 adverse events were diarrhea and neutropenia (8.7%) in the preoperative phase and thromboembolic events (5.9%) in the postoperative phase. ETS and lower SUV-2 were significantly associated with pathologic response. Conclusion The COI-B regimen is a feasible and highly active perioperative strategy in patients with molecularly unselected, potentially resectable CRCLM. ETS and SUV-2 have a promising role as imaging-based biomarkers for pathologic response.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
39
References
3
Citations
NaN
KQI