This article is more than 5 years old. Information may no longer be current.
Tumor location in colorectal cancer may predict survival
Click Here to Manage Email Alerts
The location of tumors in the colon may be an important prognostic factor for patients with treatment-naive metastatic colorectal cancer, according to study results.
Prior studies have shown tumors that appear on the right side of the colon tend to be diploid and are characterized by BRAF mutations, mucinous histology, CpG island methylation and high microsatellite instability. Left-sided tumors often are constricting, infiltrating lesions, “with a phenotype that involves chromosomal instability and aneuploidy,” according to study background.
Heinz-Josef Lenz, MD, associate director for clinical research and co-leader of the gastrointestinal cancers program at USC Norris Comprehensive Cancer Center, and colleagues evaluated the relationship between tumor location and survival in patients with previously untreated metastatic colorectal cancer who received front-line chemotherapy with or without bevacizumab (Avastin, Genentech).
The analysis included 2,027 patients from three independent cohorts. Of these, 200 were from PROVETTA, a prospective pharmacogenetic study. The others were culled from two randomized phase 3 trials, AVF2107g (n = 559) and NO16966 (n = 1,268).
The majority (70.7%) of patients had left-sided tumors.
In PROVETTA, patients with left-sided tumors demonstrated superior OS (HR = 0.44; 95% CI, 0.28-0.7) and PFS (HR = 0.52; 95% CI, 0.36-0.75) compared with those who had right-sided tumors.
Multivariable analysis in this cohort showed right-sided tumors served as a negative prognostic variable, independent of mucinous histology and BRAF mutational status.
In the two phase 3 trials, patients with left-sided tumors also demonstrated favorable outcomes. Researchers calculated HRs for OS of 0.55 (95 CI, 0.43-0.7) for patients in the AVF2107g trial and 0.71 (95% CI, 0.62-0.82) in the NO16966 trial.
The efficacy of bevacizumab was independent of tumor location, results showed.
“Given the consistency across an exploratory set and two confirmatory phase 3 studies, side of tumor origin should be considered for stratification in randomized trials,” Loupakis and colleagues wrote.
The study had several limitations, Howard S. Hochster, MD, professor of medicine and associate director for clinical sciences in the department of medicine at Yale Cancer Center, wrote in an accompanying editorial.
The analysis only included patients with metastatic disease, meaning the data may not be applicable to patients with resected tumors, Hochster said. Also, the study was a pooled analysis of three studies and did not include a control arm, meaning selection bias could have influenced the results.
“Where does this leave us?” Hochster wrote. “This interesting analysis gives rise to some important and testable biological hypotheses. The more conservative right may lean toward a platinum over a topo-1 inhibitor, and may benefit disproportionately from bevacizumab. This bears further investigation. However, at least when it comes to colon function, the right and left still work collaboratively, despite some innate differences.” – by Anthony SanFilippo
Disclosure: The researchers report employment relationships with, honoraria from, consultant/advisory roles with and grant funding from Bristol-Myers Squibb, Genentech, Regeneron, Roche and Sanofi.
Perspective
Back to Top
David H. Ilson, MD, PhD
The emergence of EGFR-targeted therapies in advanced colorectal cancer led to the discovery of RAS mutation as a predictive biomarker for the utilization of these agents. Now, with the broader and more systematic genomic characterization of colorectal cancer, it is increasingly apparent that colorectal cancer is a molecularly heterogeneous disease.In locally advanced disease, recent reports have indicated discrete molecular, pathologic and clinical characteristics. Major subtypes that have emerged include 1) tumors with more frequent presence of microsatellite instability, leading to more frequent mutation of potential tumorigenic pathways and with a more common presence of BRAF mutation, associated with a tumor location in the right or proximal colon; and 2) chromosomally unstable tumors with more frequent gene amplifications and deletions, associated with a tumor location in the left or distal colon.
Loupakis and colleagues now report results that indicate that, in metastatic colorectal cancer, right-sided primary tumors also appear to be associated with a poorer prognosis than left-sided primary tumors, and that this observation was independent of particular molecular prognostic markers. They made these observations first in a prospective pharmacogenetic study (PROVETTA) and then validated their findings in two large phase 3 clinical trials evaluating the addition of bevacizumab to chemotherapy with irinotecan plus fluorouracil/leucovorin or FOLFOX/CAPE-OX.
Tumor location (right vs. left) independently correlated with OS, PFS and antitumor response, with worse outcomes in right-sided tumors. Beneficial effects of bevacizumab were seen in both right- and left-sided colon primary site tumors. The prognostic effect of primary tumor location was maintained even when factoring in adverse prognostic factors such as BRAF mutation and mucinous histology. These results indicate that primary tumor location is prognostic and needs to be considered as a stratification variable in future trial design. They also dovetail with the divergent genomic profiles of right- and left-sided primary tumors, and underscore the need to factor in molecular characteristics of tumors in clinical trials.
Click Here to Manage Email Alerts