Navigating the Complexities of Tumor Sidedness in Colorectal Cancer

Scott Kopetz, MD, PhD, FACP

Tumor sidedness plays an integral role in the prognosis of patients with colorectal cancer (CRC), and recent research on primary tumor location has allowed practitioners to tailor treatment to improve outcomes, said Scott Kopetz, MD, PhD, FACP.

“We now have a much better knowledge of the biologic underpinning of, ‘why does right versus left matter?’” said Kopetz, an associate professor of Gastrointestinal Medical Oncology at The University of Texas MD Anderson Cancer Center. “Together, with the strong clinical data and our understanding of the biology, we are at a point where our practice should be changing.”

To date, several studies have indicated a differential response to targeted therapy in terms of tumor sidedness. Patients with right-sided CRC often have poor responses to EGFR-targeted therapies, said Kopetz, and therefore should not receive agents such as cetuximab (Erbitux) or panitumumab (Vectibix), regardless of their molecular status. In contrast, EGFR inhibition appears to improve overall survival (OS) for patients with left-sided disease.

For example, results of a retrospective analysis of the phase III CRYSTAL trial, which examined first-line FOLFIRI plus cetuximab and irinotecan in patients with RAS wild-type disease, showed that patients with left-sided tumors had a better prognosis with the EGFR inhibitor cetuximab compared with those with right-sided disease.

“In my practice, this information is indeed being utilized in the standard of care. For patients with a left-sided tumor, we are improving OS with EGFR inhibitors in the frontline setting,” added Kopetz. “I would discuss the perhaps longer duration of skin rash [with EGFR inhibitors for most patients], but for most, that improved OS is worth it.”

In an interview with OncLive, Kopetz explained the biologic difference between left- and right-sided CRC and highlighted recent research on tumor sidedness and its impact on clinical practice.

OncLive: What are the differences between left- and right-sided CRC?

Kopetz: In CRC, we've known for many years that there are differences between the left and the right side of the colon. The clinical implications of this have increasingly become apparent over the last several years. Part of the discussion is focused on what those [different] features are. What are the clinical data, and how do we actually use [these data to inform our actions] in clinic? We have known for years that the prognosis of patients with right-sided CRC is particularly poor and that patients have a shorter duration of survival and lower progression-free survival (PFS) with standard therapies. [Primary tumor location] also plays a role in recurrence after liver metastasis resection. The questions more recently have focused on how we should use this information and if there is an impact on chemotherapy.

Many studies and meta-analyses have been completed that have shown that there is differential response to targeted therapy on the right [versus the] left side of the colon. This is particularly evident with cetuximab or panitumumab. What we have seen is that patients with tumors on the right side—even though they may have a tumor that doesn't have a KRAS, NRAS, or BRAF mutation—still have a poor response to EGFR inhibitors. This would manifest with lower PFS.

The question for all of this then becomes, “Which therapies should we use for which patients?” Similarly, we have seen that when [patients with] left-sided tumors are treated with EGFR inhibition compared with a VEGF inhibitor, there are improvements in OS. For example, the CRYSTAL study showed a fairly large improvement in OS for left-sided patients who were treated with EGFR inhibitors.

The second component is really understanding why this is. Certainly, compelling randomized data are always helpful. However, for me at least, it is always good to know why that is the case. Why are we seeing a difference in biology? Here, it's important to take a step back and recognize the differences between a right- and left-sided tumor. A left-sided tumor actually derives from very different embryologic remnants of this concept of a hindgut versus midgut, back to the embryology days of medical school. It is this idea that these are very different tissues that form, and that there are many differences in the environment of the normal gastrointestinal tract. This is manifested in several different ways. We also see differences in the epigenetics, the methylation that occurs with aging normal epithelium. There are differences in the exposure to different carcinogenic bile acids in these patients.

Therefore, it's not surprising that the end result is a very different tumor. We see that the polyps are different on the right side, with a higher rate of sessile adenomas. Together, this helps explain differences in the mutation profile of the RNA-based classification.

What research is ongoing in this area?

There is continued work to try to understand how tumor sidedness integrates with several different biomarkers that we have been interested in. For example, how do we integrate sidedness into our understanding of other molecular alterations and other epigenetic features such as microRNAs that have been proposed as predictive biomarkers? We are starting to realize that sidedness is incorporating many of these biomarkers that we had been interested in in the past. It provides a "poor man's biomarker" as we like to say, where just by looking where the tumor arises, we can understand a lot of biologic features. Several ongoing studies are trying to dissect out how much of the biology can be attributed purely to the sidedness [of the tumor]. How much benefit is still present in patients who test for benefit with EGFR inhibition?

What still needs to be understood about left- versus right-sided tumors?

There are now questions about how to design studies that recognize the differences between left- and right-sided CRC. Now, we need to start asking questions within a very distinct subgroup about how to best optimize treatment. For example, within left-sided tumors, it may be that now there are biomarkers, such as the consensus molecular subtypes or HER2, that within a left-sided tumor may very clearly distinguish the potential for benefit with EGFR therapies relative to VEGF inhibition.

What is your take-home message to your colleagues?

As we understand the context of colon tumors, we have to better understand the complexities of tumor location. We then need to start to understand how to apply some of those characteristics in our understanding of the carcinogenesis of these tumors to some other major issues we are facing in the field. One of those issues is early-onset CRC; increasingly, this is occurring in rectal, low-lying colon cancer. We don't have a good understanding as to why that is. However, as we understand the normal biology of these low-lying left-sided tumors, that helps us to better define the landscape of etiology for challenges like early-onset colon cancer.

Tejpar S, Stintzing S, Ciardiello F, et al. Prognostic and predictive relevance of primary tumor location in patients with RAS wild-type metastatic colorectal cancer: retrospective analyses of the CRYSTAL and FIRE-3 trials. JAMA Oncol. 2017;3(2):194-201. doi: 10.1001/jamaoncol.2016.3797.