Regorafenib/Nivolumab Elicits Better Responses in pMMR/MSS CRC Without Liver Metastases

Although the combination of regorafenib (Stivarga) and nivolumab (Opdivo) was found to elicit responses in patients with mismatch repair–proficient (pMMR)/microsatellite stable (MSS) colorectal cancer (CRC), a subgroup of patients with liver metastases were not found to respond to the regimen, according to Marwan Fakih, MD. As such, the absence of liver metastases may be indicative of a better response with the VEGFR/PD-1 combination.

Results from a single-arm, phase 2 study (NCT04126733) presented during the 2021 ASCO Annual Meeting showed that 5 patients who did not have liver metastases at baseline experienced a durable partial response with the regimen, translating to an overall objective response rate (ORR) of 7% in all patients (n = 70).1 Moreover, 31% of all patients achieved disease stability, and the disease control rate was 39% with the doublet.

In patients without liver metastases (n = 23), the ORR achieved with the combination was 22%; the ORR in those with liver metastases (n = 47) was 0%. Moreover, the median progression-free survival was found to be longer in the patient subgroup without liver metastases vs that seen in those with liver metastases, at 3.5 months vs 1.8 months, respectively.

“The patients with liver metastases unfortunately did not respond [to the combination],” said Fakih, a professor in the Department of Medical Oncology & Therapeutics Research and the associate director for Clinical Sciences at the Comprehensive Cancer Center. “That tells us that there is a lot of work that needs to be done for [those with] liver metastases. It also tells us that maybe the combination of VEGFR inhibitors plus PD-1 inhibitors may not be as effective in patients with CRC who have liver metastases.”

In an interview with OncLive^®, Fakih, who is also the medical director of the Judy & Bernard Briskin Center for Clinical Research and co-director for the Gastrointestinal Cancer Program at the City of Hope Comprehensive Cancer Center, discussed the efficacy of regorafenib plus nivolumab in patients with pMMR/MSS CRC, the potential for VEGFR/PD-1 inhibitor combination regimens in this population, and future directions for research.

OncLive^®: Why is there an unmet need to evaluate the role of immunotherapy further in mismatch repair and MSS CRC?

Fakih: Historically, many studies have looked at the role of PD-1 targeting in patients with MSS, and the results have been disappointing. In addition, a randomized phase 2 trial [NCT02870920] looked at combining a CTLA-4 inhibitor with a PD-L1 inhibitor vs best supportive care and showed no improvement in progression free survival [PFS] or responses with [the immunotherapy combination] in this population.

As such, for the longest time, we have deemed these patients to be immunotherapy resistant, and we are trying to figure out ways to overcome that resistance. Several reports show that the combination of a VEGFR TKI [and a] PD-1 [inhibitor] may, indeed, be associated with activity.

Why is this important for this patient population? Once first- and second-line therapy [fails], they do not have good options [available to them]. The treatment options after first-, second-, and in some patients, third-line [failure]…include regorafenib alone or trifluridine/tipiracil [TAS-102] alone [depending on what was received in the first line].

Although these [agents] are important in [terms of] improving survival, the improvements are modest, and the response rates are very low, so we need better options for our patients. If we can make pMMR tumors immune responsive, and [achieve] significant benefits, then that [will be] great for our patients; it is very much needed.

What prompted the evaluation of nivolumab plus regorafenib in this population?

Approximately 2 years ago, a Japanese research group evaluated the addition of regorafenib to nivolumab in patients with MSS CRC. Some preclinical data suggested that regorafenib may have some beneficial impact as far as altering regulatory T cells in tumors, [as demonstrated in] in vivo models. Preclinically, [we have seen] some synergy between VEGF targeting, VEGFR targeting, and PD-1 targeting. [We] also [have] evidence from other tumor types that when we combine bevacizumab[Avastin], for example, with a PD-1 [inhibitor] we may also improve outcomes. [By combining] VEGFR-targeting agents with PD-1 inhibitors in renal cell carcinoma, [we saw] a high response rate.

There is plenty of clinical and preclinical rationale for targeting the VEGF pathway in combination with PD-1. The Japanese group evaluated the combination of regorafenib and nivolumab in 25 patients with metastatic CRC, 24 of whom had MSS [disease] and had initially shown a very robust response rate of 30% and a PFS exceeding 6 months. This created a buzz about 2 years about the [possibility of that]. This research prompted several groups to look at VEGFR inhibitors in combination with PD-1 inhibitors.

How do immunotherapy and TKI agents synergize? What are some of the findings reported in the phase 2 study of regorafenib plus nivolumab in MSS CRC presented during the 2021 ASCO Annual Meeting?

We have not reached a consensus as to how they really synergize fully. It is not easy to obtain serial biopsies in patients who undergo these treatments. When you do, [we are] quite limited with the amount of tissue that we get. As I mentioned earlier, at least one of the paths [forward] that is believed to be associated with this is the impact of VEGFR inhibitors on regulatory T cells, which are inhibitory to the immune response or immune activation. In addition, there may be an impact on the tumor microenvironment and other ways that may make the response to immunotherapy more favorable.

Last, VEGFA has been shown to be immunosuppressive through binding to VEGF receptors and it is possible that it may be another way [in which] this is working. The definitive pathway of synergy, frankly, is not fully elucidated, at this point. What we know is that we are seeing definitive clinical signals because the response rate with regorafenib in patients with advanced CRC is expected only to be 1%, and the response rate with nivolumab alone is expected to be nonexistent; it is rarely described and only anecdotally.

The fact that we are seeing a 7% objective response rate [ORR] in an unselected patient population with regorafenib plus nivolumab on our study, and even more importantly that we are seeing an ORR of 22% in [those with] non-liver metastases, tells us that there is some form of synergy with this combination. Interestingly, this synergy appears to be more prevalent and noticeable in patients without hepatic metastatic disease. Not only do we see synergy, but it seems like there is a clinical biomarker [that can indicate which] patients [will] respond better to this combination.

What were the key takeaways of this trial? Was there anything surprising?

The discordance in clinical responses [observed] in patients with pMMR, metastatic CRC between [those with] liver metastases and non-liver metastases was very noticeable here. This is a 70-patient clinical trial, and 23 patients did not have liver metastatic disease; 47 patients had liver metastases. We saw no responses [with the combination] in those with liver metastases. However, 5 out of 23 patients without liver metastases had an ORR, and an additional 35% had stable disease. This means that more than half of the patients had a clinical benefit [with the regimen], with a median PFS of 3.5 months in the non-liver metastases cohort [that] was [determined to be] clinically significant.

[These findings raise several questions]. Should we start thinking about targeting our patients without liver metastases with immunotherapy, and specifically with a PD-1 inhibitor plus a VEGFR inhibitor? For the sake of this study, specifically nivolumab plus regorafenib? Should we be designing different combinations for patients with liver metastases? That is [where] the field should be moving.

It is important to note that our study was with regorafenib plus nivolumab. Other combinations are being investigated, and other [studies] have looked at PD-1 inhibitors plus TKI inhibitors. Not all those studies have reported based on the pattern of metastatic disease, and we await those data because it would be important to see whether this is a class effect of the combination of PD-1 and VEGFR inhibitors, or whether this is seen specifically more with the combination of regorafenib plus nivolumab. I do suspect that it may, to a certain point, be somewhat of a class effect. However, this should be confirmed through other studies.

That is basically where we are right now. We do see a signal of benefit in pMMR non-liver metastases that we believe is clinically significant, although it is not as robust as described in the Japanese data. The reason for that is unclear. It could be related to the pattern of metastatic disease, or it could be the burden of disease. Regardless, a lot of work still needs to be done in this area.

How are immunotherapy/TKIs moving forward in the paradigm?

There is a study of lenvatinib [Lenvima] plus pembrolizumab [Keytruda]. Some studies have also been conducted in China, [that have combined] fruquintinib [Elunate] with a PD-1 inhibitor. The good thing is that whatever has been reported so far, has been [found to be] associated with clinical responses. Not every single study stratified like we did, by liver metastases vs non-liver metastases to see whether there is a clinical subgroup who would respond more than others.

We have previously reported on 95 patients with MSS CRC who were treated with immunotherapy [and] findings were similar to what we had reported during the 2021 ASCO Annual Meeting. [We saw a] lack of benefit [with the regimen] in patients with liver metastases. Clinically, to the medical oncology community, there is some hope that our pMMR patients will benefit from such combinations.

Right now, the question remains as to what the degree of benefit is and which patients should be elected for this treatment? We cannot say we have a winner combination among what has been reported so far. The reason I say that is because even with regorafenib and nivolumab, we have seen a huge discrepancy in response rates [with regard to] what has been reported in Japan and what we reported in the United States.

We can see that patient selection does matter, the burden of disease probably does matter, oligometastatic disease does matter, and liver metastases probably does matter. Those are not the same across the different studies [that have been done]. As such, [we are unable] to say that one combination is superior to another. Moving forward, we have to start thinking about randomized studies more and more so that [we can] better discern what is predicted, what the impact of therapy is, and what is related to a patient prognosis.

Reference

1. Fakih M, Raghav KPS, Chang DZ, et al. Single-arm, phase 2 study of regorafenib plus nivolumab in patients with mismatch repair-proficient (pMMR)/microsatellite stable (MSS) colorectal cancer (CRC). J Clin Oncol. 2021;39(suppl 15):3560. doi:10.1200/JCO.2021.39.15_suppl.3560