Nivolumab Plus SOC Misses PFS End Point of CheckMate-9X8 Trial, But Provides Notable Benefit in Select mCRC Subsets

The addition of nivolumab (Opdivo) to the standard-of-care (SOC) regimen of modified FOLFOX6 (mFOLFOX6) and bevacizumab (Avastin) did not improve progression-free survival (PFS) vs SOC alone in the frontline treatment of patients with metastatic colorectal cancer (mCRC), but the addition of the immunotherapy did produce encouraging efficacy in subsets of patients with baseline CMS1 and CMS3 status, according to data from the phase 2/3 CheckMate-9X8 trial (NCT03414983).¹

Results, which were presented during the 2022 Gastrointestinal Cancers Symposium, showed that the median PFS with nivolumab plus SOC (n = 127) was 11.9 months (95% CI, 8.9-15.7); the median PFS with SOC alone (n = 68) was also 11.9 months (95% CI, 10.1-12.2), equating to a hazard ratio of 0.81 (95% CI, 0.53-1.23; P = .3), and missing the primary end point of the trial. However, PFS rates after 1 year were noted to be higher with the nivolumab combination vs SOC alone.

Moreover, nivolumab plus SOC elicited an objective response rate (ORR) of 60% (95% CI, 51.0%-68.0%) vs 46% (95% CI, 33.5%-58.0%) with SOC. The median duration of response (DOR) in the investigative arm was 12.9 months (95% CI, 9.0-13.1) vs 9.3 months (95% CI, 7.5-11.3) in the control arm.

“When we looked at exploratory end points, such as biomarkers, [specifically] the CD8 percentage of cells in the tumor at the baseline or the CMS classification, we found something very interesting,” Heinz-Josef Lenz, MD, lead study author, associate director of Clinical Research, J. Terrence Lanni Chair in Gastrointestinal Cancer Research, co-director at the University of Southern California (USC) Center for Molecular Pathway and Drug Discovery said. “The benefit from the PFS curve separation becomes clear when looking at those with CMS1 and CMS3. [Patients classified as] CMS1 are the microsatellite instability–high [MSI-H] group, so we would expect a [potential] benefit, but CMS3 is the same. As such, we certainly have new insights that a subset of patients who have microsatellite stable [MSS] disease benefit [from] the combination of nivolumab and SOC chemotherapy with bevacizumab.”

In an interview with OncLive^®, Lenz, who is also a professor of medicine and preventive medicine, co-director of Colorectal Cancer, and co-director of USC Norris Center for Cancer Drug Development; co-leader of the Gastrointestinal Cancers Program at USC Norris Comprehensive Cancer Center; and section head of GI Oncology in the Division of Medical Oncology at the Keck School of Medicine of USC, discussed the safety and efficacy of nivolumab plus SOC in patients with mCRC, shared additional insight from the CheckMate-9X8 trial, and alluded to next steps with further exploring the regimen in specific subsets who appear to derive benefit from this approach.

OncLive^®: What was the rationale for the CheckMate-9X8 trial?

Lenz: It is known that chemotherapy targeted agents have efficacy in metastatic disease and are considered SOC. Data from different diseases and different studies [suggest] that the addition of immune checkpoint inhibitors can increase the efficacy of SOC treatments. [CheckMate-9X8] is a randomized study using SOC, in this case mFOLFOX6 with bevacizumab, plus or minus nivolumab.

Could you elaborate on the design of the study? What key end points were evaluated?

This is a phase 2/3 clinical trial, and the recruitment was 195 patients. It was a 2:1 randomization, so 2 out of 3 patients received nivolumab in combination with SOC. The [primary] end point [of the trial] was PFS. There was also some stratification based on tumor location to better understand the potential prognostic value of the tumor [sidedness].

What were the key findings?

Overall, the median PFS [between the arms] was the same; there was no difference. However, when you look at the PFS, something interesting is happening. After 12 months, the curves separate, and the nivolumab combination with SOC flattens out. When looking at 15 months, 45.0% was the PFS rate in the nivolumab arm, vs only 21.5% in the SOC [arm]. Even later, at 18 months, [the rates are] 28.0% vs 9.0%, [respectively]. The curves separate. The median [follow-up] does not capture it. Since patients with MSS and MSI-H disease were recruited to the trial, [it was thought] that [maybe] this difference is driven by the MSI-H tumors, but it was not. The patients [with MSS disease] have the same pattern of separation after 12 months.

Beyond the primary end points, there were also secondary end points. For example, the response rate was higher in the nivolumab combination arm [vs the SOC-alone arm], and the DOR was also much longer.

We also looked at the CD8 immunohistochemistry, and [patients with a baseline CD8 level of 2% or greater] also had a PFS benefit. The problem is, overall, the study did not meet its end point, and that is 1 of the reasons that it did not move into a phase 3. But it creates a new insight in understanding that a subset of this patient population benefits from this treatment, and they may be characterized with the CMS classification or the immune cell trafficking via, in this case, CD8. We need for more data to better understand [which] patients would benefit the most [from this approach].

Did it come as a surprise that comparable benefit was observed between the MSS and MSI-H populations?

Yes, I was very surprised. There are very inconclusive data when you combine immunotherapy with chemotherapy. Sometimes [this approach] seems to help, [such as] in gastric [cancers], and in others, it does not seem to help at all. It is maybe not so surprising [with MSI-H], but it is surprising that we see a subset in MSS. It is not the majority [of the patients], which is the reason the median does not change, but it is a clearly defined subset with the separation of the curves.

That is something we did not expect because the immune-activated pathway we expect in CMS1 and CMS4, not CMS3. We do not know much about CMS3 because most of our recent studies have been looking at patients who [have] RAS wild-type [disease], and we know KRAS-mutant is in CMS3. Most of the big studies did not include KRAS-mutant cells. The whole association of CMS and outcome is usually done in RAS wild-type. The CMS3 group is very small. This is exciting and opening new ways to look at this patient population and potentially find better molecularly defined groups who benefit from this combination.

How did the addition of nivolumab to SOC impact the tolerability of the regimen?

No new safety signals [were observed]. However, numerically, [we saw] more grade 3 and 4 toxicities in the combination arm with nivolumab, as well as more dose delays. However, [we did] not see anything unexpected.

Beyond taking a deeper look at this combination in this new subset, is there anything else that you would like to see further evaluated?

The [data regarding the] CD8 cells were somewhat expected. However, the CMS3 data are amazing. We would expect [what we saw with] CMS1, but not [with] CMS3. [What we found]will initiate a lot of digging into the molecular makeup of this specificgroup.

Is there anything else about this study that you would like to emphasize?

The exploratory biomarkers are the exciting part [of this research]. There was no difference in PFS [between the arms]. The follow-up of 21 months is still short in terms of overall survival. We need to follow [these patients] longer, if these curves also separate like the PFS [curves do].

Reference

1. Lenz H-J, Parikh AR, Spigel DR, et al. Nivolumab (NIVO) + 5-fluorouracil/leucovorin/oxaliplatin (mFOLFOX6)/bevacizumab (BEV) versus mFOLFOX6/BEV for first-line (1L) treatment of metastatic colorectal cancer (mCRC): phase 2 results from CheckMate 9X8. J Clin Oncol. 2022;40(suppl 4):008. doi:10.1200/JCO.2022.40.4_suppl.008