2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Early identification of patients with liver-only metastatic colorectal cancer who are eligible for resection is crucial to determining their course of treatment and what role modalities like chemotherapy will play in their treatment strategy.
Early identification of patients with liver-only metastatic colorectal cancer (CRC) who are eligible for resection is crucial to determining their course of treatment and what role modalities like chemotherapy will play in their treatment strategy, according to Patrick Boland, MD.
“We need to consider patients who are initially resectable vs patients who are not initially resectable differently, because the goals are different,” Boland said. “For patients who are initially resectable, the goal of chemotherapy is to reduce the risk of recurrence, [or possibly] to biologically select patients who should undergo surgery. For patients who are not resectable upfront, the goal is to shrink the tumor as much as we can to enable surgery. [This] leads us to think about more aggressive treatments.”
In an interview with OncLive®, Boland, a medical oncologist, member of the gastrointestinal oncology team, and active member of the National Cancer Institute’s National Clinical Trial Network, discussed the evolution of treatment for patients with liver-only metastatic CRC and the different treatment options for those who are initially resectable vs not initially resectable.
Boland: There is a debate [about] what chemotherapy adds [to treatment]. Right now it is the standard. When [considering] patients as a whole, we have to think about those who are initially resectable upfront, or those who are not resectable but are convertible with chemotherapy, or at least [who are at] risk for positive margins.
[For patients] with resectable disease, the EPOCH trial [NCT01483027] and the phase 2/3 JCOG0603 trial both [demonstrated] a decrease in rates of recurrence with the use of perioperative 5-fluorouracil, leucovorin, and oxaliplatin [FOLFOX]. However, [neither trial] demonstrated a survival benefit. We can look at that a few different ways, but one of the major goals of giving chemotherapy is to increase the rate of curing patients and decrease risk of recurrence. The gold standard clearly is survival, and that is what we would love to show. The fact that patients can go on to be salvaged with other therapies proves that the bar is perhaps too high for any of these studies to show individually.
Right now, perioperative chemotherapy remains a standard, with FOLFOX or capecitabine plus oxaliplatin [CAPOX] being the [usual choice]. There have been other studies looking at adding targeted drugs in resectable disease, but to date, they have failed to show any improvement. For instance, cetuximab [Erbitux] failed to show improvement, so we do not recommend that in resectable disease upfront.
On the other hand, with disease that is not resectable initially, the options for therapies keep expanding. We have seen an increased use of triplet chemotherapy, such as FOLFOXIRI or FOLFIRINOX, which gives very high response rates. We also have some data that adding bevacizumab [Avastin] to FOLFOX improves response rates, and [might] protect the liver. The good news is that as we learn how to better use therapies, we have more options. The tricky part is figuring out how to use the right amount at the right time.
In the phase 2 OLIVIA trial [NCT00778102], patients received FOLFOXIRI and bevacizumab and we saw that there is a higher rate of R0 resection with the use of triplet chemotherapy vs doublet chemotherapy. We do not know if that translates to improved long-term outcomes yet. However, if you believe that getting patients to surgery gives them the best long-term odds, then certainly for patients that are not resectable initially, and who are fit enough to [receive] triplet chemotherapy, this represents a feasible option.
There is also Japanese data looking at which biologics should be used if we are giving triplet therapy to patients who are RAS wild-type. Is bevacizumab better? Or is cetuximab better? [Based on] the DEEPER study (JACCRO CC-13; UMIN000018217) that was presented at the 2021 ASCO Annual Meeting, we see that both have similar response rates, which [might be] surprising. We get a little bit of a deeper response with the anti-EGFR drugs, but the number of patients who undergo high-quality resections did not differ. There is not necessarily a preferred [agent] in that space. I tend to lean towards bevacizumab, but we can look at this on a case-by-case basis based on how much response a patient needs and who the patient is.
The first thing we look at is how much disease there is in the liver, and then any other risk factors. Does the patient have synchronous or metachronous presentation? Do we know anything about mutational status or poor prognosticators, such as whether the patient is RAS-mutated, or BRAF-mutated? Do we know anything about the primary [tumor], and is it intact? Was it resected or node positive?
Bilobar disease, metastases, and mutations are all bad prognosticators [and patients with these characteristics are more likely to be] given chemotherapy upfront, even if they are resectable. We do not know that that is the best way to get long-term outcomes, but in general, our practice is commonly to give perioperative chemotherapy upfront.
We try to limit [perioperative chemotherapy] as much as possible. The aim is to give no more than 2 to 3 months if the patient is resectable at that time. [We give patients] frequent scans every 6 to 8 weeks to reassess. This always requires an upfront visit with a hepatobiliary surgeon, which needs to be on the radar when patients have liver-only metastases. Even if you [suspect] the [patient is] not resectable, it is worth a discussion whether portal vein embolization or some other out-of-the-box techniques can be employed.
The new kid on the block, and what represents the future is circulating tumor DNA [ctDNA]. We have seen this in the gastrointestinal space and across [other] cancers. There are multiple studies that are emerging now showing that the presence of ctDNA after surgery is a poor prognosticator, whereas its absence is a good prognosticator.
Optimally, we need randomized studies completed to show us what to do with that [information and] how to make the most of it. Can we omit chemotherapy for patients who are negative since we do not have a well-proven survival benefit with postoperative chemotherapy? For patients who are positive, do we have room to escalate therapy? Will that improve long-term outcomes? There are studies that are being developed now that hopefully will answer those questions.