BARCELONA, Spain—The evidence from several clinical trials and laboratories is undeniable—colorectal cancer in the right side of the colon is a different disease from colon cancer in the left colon.
“Again and again we see that right-sided colon cancer is a different disease, and in the future it should be treated in different ways,” said Heinz-Josef Lenz, MD, Professor of Medicine and Preventive Medicine and Director of the Gastrointestinal Oncology Program at the USC Norris Comprehensive Cancer Center, Los Angeles, in a presentation at the recent European Society of Medical Oncology (ESMO) World Congress on Gastrointestinal Cancer
Two Sides, Different Cancers
Right-sided colorectal tumors arise from the mid gut and left-sided tumors from the end gut, Lenz said, accounting for two distinct embryological tissues that lead to two different cancers. There are additional differences as the cancer develops, so that clinically they behave differently, there are different environmental data, and the microbiota is completely different left from right. All this leads to a distinct difference in molecular genetics.
“The prognosis in colorectal cancer seems not to be related to the prognostic markers we all know, like MSI (microsatellite instability) and BRAF, or to the CMS (consensus molecular subtypes) classification, or methylation status,” Lenz said. “But early data show that sidedness may be predictive for targeted agents and maybe even for chemotherapy.”
Tumors on the right side typically have DNA mutations including BRAF, PIK3CA kinase, and KRAS mutations, and more MSI. This high mutation frequency points to a poor prognosis.
And the right-sided primary tumor is also associated with CMS 1 and 2, Lenz noted. “But you also see more undifferentiated cancer on the right side, which is very important for the distinction of benefit of target agents.”
Tumors on the left side have more chromosomal alterations—18q, 20q, 22q—but also, very importantly, they have a significant upregulation of the HER pathway, not only for HER1 and the EFGR receptor but also HER2.
“When you see HER2 amplification in colon cancer, it's usually from a left-sided colon cancer,” Lenz said. “But what we really need to know is whether these classifications change or explain the differences in outcome.”
Different Treatment for the Right Side
Differences in outcome were clearly seen in the phase III Southwest Oncology Group (SWOG) 80405 trial, one of the reports highlighted at the 2016 ASCO Annual Meeting (Abstract LBA3).
A detailed retrospective analysis of SWOG 80405 confirmed what had been suspected from earlier clinical trials—tumors of the left side and right side responded differently to the same drugs, and right-sided colorectal cancer is a different disease than left-sided colorectal cancer.
Patients in SWOG 80405 received either bevacizumab or cetuximab, both with standard chemotherapy. Overall survival showed no statistically significant difference between the two regimens. But median overall survival among all patients with left-sided tumors was 33.3 months, versus 19.4 for right-sided tumors. The difference was also significant in progression-free survival: 11.7 months for left-sided tumors versus 8.9 months for right-sided tumors.
“And survival outcomes in patients with KRAS wild-type metastatic colorectal cancer were significantly longer among those with tumors originating on the left versus the right side of the colon,” Lenz said.
More important was the difference in overall survival by targeted therapy. Median overall survival with cetuximab was 36.0 months for left-sided tumor versus 16.7 months for right-sided tumors.
“The 19.3-month difference is a big difference—sidedness is prognostic,” Lenz explained.
The difference with bevacizumab was not statistically significant, 31.4 months versus 24.2 months, respectively. But those data were independent of tumor biology, protocol therapy, prior adjuvant chemo- and/or radiotherapy, age, and sex.
“This all points out that for right-sided colon cancer we may have to go back to the drawing board” to find an effective treatment, he said.
Lenz said earlier there was a hint of this sidedness issue in the National Cancer Institute of Canada (NCIC) CO.17 clinical trial (Eur J Cancer 2015;51:1405-1414). That trial showed the effect of the epidermal growth factor receptor (EGFR) inhibitor cetuximab was strongest in the left-sided colon cancer. “This makes biological sense because this is where the pathway is activated.”
In that trial, there was also a hint that response to cetuximab may be not only prognostic but predictive as well for targeted agents.
Lenz said colorectal cancers differ at the molecular level not only by location but also by gender. He cited unpublished gender data from his group based on The Cancer Genome Atlas.
“The difference in males between the right and left colon is 148 genes, and for women the difference is 2,371 genes,” he noted.
The number of genes is not the most critical difference between colon cancer in males and females, but rather it is the differential pathways activated, according to Lenz.
“Since these right-sided tumors often have no expression of the EGF receptor or ligand, can we differentiate and then treat with different agents? Research in this area with patient-derived xenograft models is ongoing,” Lenz said.
Because many tumors on the right side are methylated, Lenz and colleagues looked in two cell lines, Smad interacting protein (SIP)-positive and SIP-negative, to see if they would change with demethylation.
“We find we can get increased apoptosis treating with 5-azacytidine combined with vitamin C,” he said. “Now we are planning to do demethylation tests in combination with targeted agents and immuno checkpoints.”
Lenz noted specimens from several clinical studies now being analyzed for molecular differences: for EGFR-versus-bevacizumab studies the trials are SWOG 80405, FIRE-3, and PEAK; and for EGFR-versus-no-EGFR the trials are PRIME, CRYSTAL, OPUS, COIN, NORDIC VII, and the New EPIC trial.
“In the future, do not consider just the typical markers of age, performance status, stage, and the typical molecular markers RAS, BRAF, and MSI [when choosing a therapy for colon cancer], but also consider which side the tumor is on, until we know what is the surrogate for outcome,” Lenz concluded.
“We need to find better treatments for right-side colon cancer because the benefits of any targeted agents are very little.”
Robert H. Carlson is a contributing writer.