Third-Line Treatment Strategy in mCRC Shifts Focus From Tumor Response to Disease Stabilization

With the extremely low response rates often seen with FDA-approved third-line regimens in metastatic colorectal cancer (mCRC), therapeutic intent must shift away from tumor shrinkage and toward durable disease control; likewise, treatment selection in this setting must prioritize balancing patient characteristics and prior toxicities with potential therapeutic benefit and quality of life, according to Benjamin A. Weinberg, MD, FACP.

“[Achieving this] balance can sometimes feel like walking a tightrope, as it can shift even day to day or hour to hour,” Weinberg, an associate professor of medicine in the Division of Hematology and Oncology at the Lombardi Comprehensive Cancer Center at Georgetown University in Washington, DC, shared in an interview with OncLive^®. “But when choosing between third-line systemic options, we have to be conscious of patient factors, especially for those who have had issues with cytopenias during prior cytotoxic chemotherapy.”

In the interview, Weinberg highlighted the importance of patient-specific factors, prior toxicities, and up-front or repeat biomarker testing for making informed treatment decisions; discussed new immunotherapy combinations and targeted therapies potentially moving into earlier lines of treatment; and stressed the need for proactive treatment escalation to support the shift toward less aggressive approaches in the adjuvant setting.

OncLive: How do you define the primary therapeutic goal in the third-line setting for patients with mCRC, and how does that goal differ from earlier lines of treatment?

Weinberg: In the third-line setting, we are very cautiously balancing the toxicity of our treatment against the therapeutic benefit. We know that many of the approved drugs in the third-line setting have response rates close to 0%, so we are, at best, hoping to achieve stabilization without causing too much toxicity in the process. This is also a ripe setting for new treatments and clinical trials; much of what we do is test new drugs in the third-line setting, as opposed to waiting until [a patient] has exhausted all standard therapies, which now include third-, fourth-, and even fifth-line indications.

Once patients have progressed on standard first- and second-line regimens, what clinical and molecular factors most strongly influence your approach to third-line sequencing in mCRC?

In the third line, we generally have 4 standard-of-care [SOC] systemic options. One is to rechallenge with an oxaliplatin-based chemotherapy. These are often patients who benefited from frontline FOLFOX [leucovorin, fluorouracil (5-FU), oxaliplatin], CAPOX [capecitabine (Xeloda), oxaliplatin], or FOLFIRINOX [leucovorin, 5-FU, irinotecan, oxaliplatin] [for whom] we then discontinued oxaliplatin due to concerns about potential peripheral neuropathy. In patients without neuropathy who benefited in the first-line setting, it is not unreasonable to rechallenge with a FOLFOX or CAPOX combination in this setting, especially if we are trying to [bolster] response.

The other SOC options include trifluridine/tipiracil [Lonsurf], with or without bevacizumab [Avastin], and regorafenib [Stivarga]. Again, these treatment options have very limited response rates, and we often make a decision based on prior toxicities and patient characteristics. For example, a patient who has had many issues with cytopenias, such as anemia and neutropenia, might find it difficult to receive trifluridine/tipiracil. For those with uncontrolled hypertension or concerns regarding potential bleeding or even perforation risks, we would generally not add bevacizumab to trifluridine/tipiracil and we would also generally shy away from fruquintinib or regorafenib because of the potential VEGF-related adverse effects [AEs]. Those are mostly the things we keep in mind when trying to decide which third-line systemic therapy to choose.

How do key biomarkers such as RAS, BRAF, microsatellite instability (MSI) status, HER2 amplification, and tumor sidedness shape third-line treatment selection, and when do you consider repeat molecular testing or liquid biopsy in this setting?

I would argue that biomarker-driven therapy in mCRC should take place in the first-line setting. We now know that if a patient is MSI-high or mismatch repair deficient [dMMR], they should receive immunotherapy up front. If they are RAS/BRAF wild-type, I would argue—especially if they have a left-sided primary tumor—that they should receive EGFR therapy with chemotherapy up front.

HER2 is a bit of a different character because it does have a later-line indication. This is a place where you really want tissue testing that shows immunohistochemistry [IHC] 3+; you can repeat molecular testing on biopsies or even consider a liquid biopsy in that setting. However, I would be cautious. I believe the only benefits for RAS wild-type tissue that is HER2 [IHC] 3+ are [from] the tucatinib [Tukysa] and trastuzumab [Herceptin] combination. Fam-trastuzumab deruxtecan-nxki [Enhertu] could be considered as well, especially if there is a concurrent KRAS mutation.

We often repeat molecular testing if it has been a long time since the up-front molecular testing [was conducted]. BRAF is another good example where, if someone has not been exposed to BRAF-targeted therapy in the first or second line for some reason, we would want them to have access to encorafenib [Braftovi] with an EGFR antibody in a later-line setting. It is important to repeat molecular testing often, especially in this era of liquid biopsies, where we can get results back quite quickly. But for the most part, much of the biomarker selection has probably already taken place by the time we hit the third line.

With multiple later-line options in mCRC supported by only modest efficacy data, how do you translate those findings into real-world treatment decisions?

We always try to shy away from cross-trial comparisons, but when multiple agents are approved in the later-line setting [with the comparator being] placebo, it's hard not to, especially with similar patient populations. We know that trifluridine/tipiracil plus bevacizumab is superior to trifluridine/tipiracil alone. As a result, that regimen is probably, at least for now, cemented as the preferred third-line regimen in patients without preexisting cytopenias or other relative contraindications. Based on the [phase 3] FRESCO-2 [NCT04322539] data, we do know that following fruquintinib [Fruzaqla] after regorafenib doesn't make a whole lot of clinical or biologic sense.¹ Nowadays, all things being equal, I would probably move fruquintinib up to a fourth-line setting if it can be covered by insurance, which has been an issue even in later-line settings, because it's quite expensive. Oftentimes, insurance companies want patients who have been exposed to every prior treatment before getting that drug. Those are the things we [consider], but it is hard to do these cross-trial comparisons because they were different patient populations.

How do cumulative toxicities from earlier lines of therapy factor into third-line decision-making, and how do you balance potential benefit with quality-of-life considerations at this stage of disease?

It is important to be very mindful of potential toxicities of the treatments we are offering. For patients with difficult-to-manage neutropenia or anemia, trifluridine/tipiracil is not an easy drug to give if they are already at that baseline. Similarly, the VEGF-targeted therapies—whether adding bevacizumab to trifluridine/tipiracil or using regorafenib or fruquintinib—would be difficult for populations with uncontrolled hypertension. The same applies to bleeding risk; it can be quite hard to use those medications in patients who have a primary tumor in place that is bleeding. We are also concerned about VEGF therapy for patients at increased risk for AEs such as perforation and even thromboembolic disease. In the absence of any of those, trifluridine/tipiracil with bevacizumab would still be the preferred choice, but we take those prior AEs into very high consideration when making this treatment selection.

How do you expect the third-line mCRC landscape to evolve as targeted therapies and biomarker-driven strategies continue to move earlier in the treatment paradigm?

Based on data from the phase 3 STELLAR-303 trial [NCT05425940], the newest combination on the block is zanzalintinib [XL092]—an oral multi-TKI [tyrosine kinase inhibitor] that hits VEGF among other targets—plus the anti–PD-1 agent atezolizumab [Tecentriq]. That [regimen] did show a significant overall survival benefit vs regorafenib [in patients with relapsed/refractory] mCRC that was not MSI high or dMMR, although the improvement was slight and the toxicity is real; it is currently awaiting FDA evaluation.² If that [regimen] makes it into the treatment landscape, there may be patients we might steer more toward that over trifluridine/tipiracil plus bevacizumab in certain circumstances. For [a patient] who has been rapidly progressing on chemotherapy, fluoropyrimidine-based therapy, and VEGF therapy, it just makes more sense, both clinically and biologically, to try something different, which would be immunotherapy.

Ideally, this type of treatment would move into earlier lines of therapy, potentially in the maintenance setting, where we think patients can derive more benefit from immunotherapy. Because it takes longer to derive benefit, saving it for the third line may not be the optimal population. [Zanzalintinib plus atezolizumab] also seems to have activity in [patients with] liver metastases, whereas other novel immunotherapy combinations, like botensilimab/balstilimab, do not seem to have as much effect in liver metastases. That is a specific population where it will be interesting to see how these new immunotherapy combinations shake out. For the most part, I do not know if any of them are going to supplant our current treatment paradigm yet. Biomarker-driven strategies are important to try to identify—outside of the liver vs non–liver metastasis populations—which biomarkers predict benefit from TKI/immunotherapy combinations, because other than liver metastases, we do not really have one.

References

1. Dasari A, Lonardi S, Garcia-Carbonero R, et al; FRESCO-2 Study Investigators. Fruquintinib versus placebo in patients with refractory metastatic colorectal cancer (FRESCO-2): an international, multicentre, randomised, double-blind, phase 3 study. Lancet. 2023;402(10395):41-53. doi:10.1016/S0140-6736(23)00772-9
2. Hecht JR, Park YS, Tabernero J, et al; STELLAR-303 Study Investigators. Zanzalintinib plus atezolizumab versus regorafenib in refractory colorectal cancer (STELLAR-303): a randomised, open-label, phase 3 trial. Lancet. 2025;406(10517):2360-2370. doi:10.1016/S0140-6736(25)02025-2