What are the Best Late-Line Metastatic Colorectal Cancer Treatment Options?

The later-line metastatic colorectal cancer (CRC) space has many treatments available to patients. Among them are trifluridine/tipiracil (Lonsurf) plus bevacizumab (Avastin), as seen in the phase 3 SUNLIGHT trial (NCT04737187), and fruquintinib (Fruzaqla) monotherapy, per the phase 3 FRESCO-2 trial (NCT04322539). In SUNLIGHT, trifluridine/tipiracil plus bevacizumab demonstrated favorable overall survival compared with trifluridine/tipiracil alone in patients with refractory metastatic CRC who received no more than 2 prior chemotherapy regimens.¹ In FRESCO-2, fruquintinib improved OS vs placebo in patients with refractory metastatic CRC who had progressed on or were intolerant to trifluridine/tipiracil or regorafenib (Stivarga).²

With these options available, physicians are faced with many data points that need to be interpreted in accordance with the patient’s treatment goals, whether that’s a patient not wanting to receive infusions anymore or just wanting to achieve the best possible disease control.

CancerNetwork® spoke with Marwan G. Fakih, MD, about the later-line metastatic CRC treatment landscape during the beginning of Colorectal Cancer Awareness Month. Fakih spoke about all the considerations he faces when treating a patient with late-line CRC. The conversation also touched upon a recently published paper he coauthored that revealed a correlation between the occurrence of severe neutropenia and improved outcomes with the SUNLIGHT regimen.³

Fakih is a professor in the department of Medical Oncology and Therapeutics Research, the deputy director of City of Hope Comprehensive Cancer Center, medical director for Briskin Center for Clinical Research, and division chief of Gastrointestinal Medical Oncology and co-director of the Gastrointestinal Cancer Program at City of Hope.

CancerNetwork: What does the later-line metastatic CRC treatment landscape look like right now?

Fakih: For non-biomarker driven colorectal cancer where there are no associated, clear targeted therapies, our options today are basically consideration for [trifluridine-tipiracil] plus bevacizumab as a treatment option, fruquintinib as a treatment option, as well as regorafenib for patients who are not candidates for any anti–VEGF or anti–VEGFR targeted therapies. Trifluridine monotherapy is also considered a treatment option for advanced metastatic colorectal cancer. As a provider and or as a treating physician, we have to make some decisions here when a patient progresses on oxaliplatin, 5-[fluorouracil], irinotecan, plus bevacizumab, and anti–EGFR if indicated. The data is solid for each of these agents, in that they improve overall survival, but there are certainly some differences in progression-free survival, and disease control, as well as some toxicities that are unique to each regimen.

How do you differentiate between a patient who is a better candidate for the SUNLIGHT or FRESCO-2 regimen?

The first thing that comes to mind when we’re seeing a patient in a third-line setting is, “What is the treatment that is most likely to control the disease the longest, and what is the treatment that has been associated with the longest impact on overall survival [OS]?” In my practice, I extrapolate a lot from SUNLIGHT, and that’s a randomized phase 3 trial that looked at [trifluridine/tipiracil] plus bevacizumab, in comparison, not to placebo, but rather to an active control that was approved as a standard of care in a third line setting, which is [trifluridine/tipiracil alone]. In that trial, we have seen a substantial improvement in both progression-free survival [PFS] and OS, where you have a median PFS of 5.6 months for [trifluridine/tipiracil] and bevacizumab. That is a clinically meaningful time point and end point, in my opinion. This has translated on that clinical trial in a substantial improvement in OS over [trifluridine/tipiracil] monotherapy, and has resulted in more than a 3-month improvement in OS. Half of the patients in that advanced setting were alive beyond the 10-month mark. When we compare that with other studies—and one should not do cross-study comparisons—the use of regorafenib and [trifluridine/tipiracil] alone had diminished substantially because the median PFS is approximately 2 months. That means half of your patients will progress on their first scan.

There’s no doubt in my mind that [trifluridine/tipiracil] plus bevacizumab has a substantial edge from the perspective of PFS in our patients, and that it can control the disease longer, which also improves the quality of life of patients, in that there is a delayed deterioration in PFS and quality of life when you when you control the disease. That's what we’ve seen on SUNLIGHT. The sunlight trial showed a delay in deterioration in ECOG performance status, so for most of my patients, I choose [trifluridine/tipiracil] plus bevacizumab.

Recently, we have seen the approval of fruquintinib as well, based on the FRESCO-2 trial, and that’s an active agent. The median PFS was 3.7 months, but still, we are dealing with a median OS in the 7-month range. It’s important to note that FRESCO-2 was designed so that patients should have progressed on [trifluridine/tipiracil] or regorafenib before they received fruquintinib. Our data from FRESCO-2 is more substantial for fourth-line therapy. How do I decide on treatments? In this somewhat “confusing” landscape today, I let the data drive my decision-making. Unless there is a contraindication for [trifluridine/tipiracil] and bevacizumab, that would be my go-to regimen in a third-line setting.

Was there a patient groups that derived more or less benefit from trifluridine/tipiracil plus bevacizumab?

No. Everybody benefited, including patients who had prior bevacizumab. There was no impact of RAS status on outcomes, and even patients with liver metastases benefited from the addition of bevacizumab. Notably, there has been a little bit more improvement in terms of hazard ratio for the patients who were bevacizumab naive. That’s not to be expected for a patient who did not have bevacizumab before. The impact of bevacizumab may be more intense or more significant, but if you look at the hazard ratios for patients who received bevacizumab prior to receiving [trifluridine/tipiracil] plus bevacizumab, the data remained solid. There was a statistical improvement in PFS in the group that received bevacizumab. If we were to look at what is roughly the median PFS for patients who receive prior bevacizumab, you’re still expecting somewhere in the neighborhood of between 4 and 4.5 months in patients who received prior bevacizumab. If you look at patients who had no prior bevacizumab, then the median PFS well exceeds 6 months. Yes, there is a difference, but everybody benefits. There are no patients who should be excluded from receiving trifluridine/tipiracil plus bevacizumab in a third-line setting based on SUNLIGHT.

How do you aid in your patient’s cancer treatment decision-making?

It’s important to guide patients in your decision-making as well. Patients are looking at their physicians, seeking guidance. There are pros and cons for every strategy; the most important outcome is going to be eventually attained by making sure that all active agents are offered to your patients, and by also being mindful of what the patients' priorities and goals are. Those could be variable from one patient to another. Some patients may say, “I don’t want to come for infusions anymore. I just want to take an oral medication.” That may be somebody who should be getting fruquintinib. Some patients, on the other hand, may be asking the question, “Which regimen has the highest disease control? What are the best chances that I maintain a PFS end point at the 4-month mark?” Numerically speaking, if we look at SUNLIGHT, that's mostly with [trifluridine/tipiracil] and bevacizumab.

Quality of life is important. A lot of people flag [trifluridine/tipiracil] as an agent that is associated with bone marrow suppression, which is true; however, the reality is that neutropenia and leukopenia are blood tests, not a bad feeling. If your white [blood cell] count is low, that doesn’t mean you’re having severe fatigue. The reassuring thing about SUNLIGHT is that the rates of febrile neutropenia were very low on the [trifluridine/tipiracil] plus bevacizumab arm. Really patients had an outside the bone marrow suppression and low rates of severe toxicities. We’re privileged to have the option of using G-CSF support and maintaining treatment intensity in patients who are tolerating the treatment well otherwise. I have not seen resistance to using [trifluridine/tipiracil] plus bevacizumab map from my patients, and I've seen some fairly good results in disease control in patients who receive [this regimen] in the third-line setting. Are there patients who I think are better served with fruquintinib in that setting? If you have a patient who has a baseline platelet count of 70,000, that's a patient who may have a hard time with thrombocytopenia during treatment and may be considered for other strategies in that scenario. It’s individualized and a patient discussion, and the physicians who know their patients best and have been treating them through that journey will be able to guide them to the best approach.

Patients who received trifluridine/tipiracil plus bevacizumab experienced improved outcomes when they had severe neutropenia. What do these results mean?

The question of neutropenia and benefit from [trifluridine/tipiracil] has been asked before with [trifluridine/tipiracil] monotherapy, in that there had been data suggesting that patients who developed significant neutropenia had better results with [trifluridine/tipiracil] compared with best supportive care. In the past, there has been a lot of debate; however, we ask if it is the chicken or the egg. Is the fact that you’re having neutropenia going to imply that you’re going to get a better response? Or is it the fact that, because you responded better to therapy, you stayed longer on the treatment, and you’re getting neutropenia?

The nice thing about SUNLIGHT is that we shed some light on that. If you do get neutropenia, even on the first cycle where you haven't really used G-CSF yet, you’re more likely to derive a benefit that is deeper. That's an important finding. It doesn't mean you need to make your patients sick to get a benefit, but it means that neutropenia is a pharmacodynamic biomarker of adequate dosing. If you do get neutropenia with the standard dosing, that means that you’re probably also having a better cytotoxic effect on your cancer cells. Now, it doesn’t mean that we should consider higher dosages for those who don’t have neutropenia. That would be a no-no because no one has tested that safety strategy, but to me, what it means is that if somebody develops neutropenia and otherwise does not have any other [adverse] effects, I, as a treating physician, would favor considering G-CSF support on that individual rather than reducing the dose. The patient is tolerating the treatment well, and you can fix the neutropenia, especially when that neutropenia is a biomarker of benefit. Perhaps it tells us that dosing is important. A dosing regimen that has a pharmacodynamic end point that is notable, such as low [white blood cell] count and low [neutrophil] count, may also be important in efficacy.

How does this association between neutropenia and improved outcomes with trifluridine/tipiracil affect G-CSF usage and considerations?

We follow ASCO guidelines for most; however, if you look at SUNLIGHT and other studies, roughly about a third of patients will require G-CSF support. Now, this is a heavily pretreated patient population that has gone through first-line and second-line therapy before receiving [trifluridine/tipiracil] plus bevacizumab, which means that many of the patients that I treat in a third-line setting have been on systemic chemotherapy for a year and a half to 2 years. The bone marrow gets tired, so you have to ask if we are at a point where we should consider it prophylactically with cycle 1. Not necessarily, but, in general, a lot of patients will need it with [trifluridine/tipiracil] plus bevacizumab, and we shouldn't shy away from it if it’s important to maintain safety and efficacy. It has been adopted quite frequently in the community and in treating patients with [trifluridine/tipiracil] plus bevacizumab. In general, when I use it, I use it on day 15, which is the day of bevacizumab alone, after completing the first 2 weeks of therapy. In that setting, I almost never see a delay in treatment.

References

1. Prager GW, Taieb J, Fakih M, et al. Trifluridine-tipiracil and bevacizumab in refractory metastatic colorectal cancer. N Engl J Med. 2023;388(18):1657-1667. doi:10.1056/NEJMoa2214963
2. Dasari A, Lonardi S, Garcia-Carbonero R, et al. 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
3. Prager GW, Elez E, Fakih MG, et al. Association between neutropenia and efficacy in patients with refractory mCRC receiving trifluridine/tipiracil + bevacizumab: post hoc analysis of the SUNLIGHT trial. ESMO Gastrointest Oncol. 2025;10:100234. Published online September 8, 2025. doi:10.1016/j.esmogo.2025.100234