Immunotherapy Disappoints in Metastatic Colorectal Cancer

Liam Davenport

June 27, 2018

BARCELONA, Spain — Hope that arose from an earlier trial turned into disappointment with the results of the IMblaze370 phase 3 study, conducted in patients with metastatic colorectal cancer (mCRC) who experienced disease progression while taking chemotherapy.

The trial showed that a new approach that involved using immunotherapy with the checkpoint inhibitor atezolizumab (Tecentriq, Genentech), either alone or with the MEK1/2 inhibitor cobimetinib (Cotellic, Roche), showed no survival benefit over the tried and tested approach of using regorafenib (Stivarga, Bayer).

The study, which included more than 360 mCRC patients for whom at least two prior chemotherapy regimens had failed, showed that neither atezolizumab alone nor in combination with cobimetinib yileded an improvement in overall survival, objective responses, or progression-free survival when compared to regorafenib.

The new data were presented here at theWorld Congress on Gastrointestinal Cancer 2018 by Johanna Bendell, MD, Sarah Cannon Research Institute/Tennessee Oncology, Nashville.

The "lack of clinical activity may be due to immune-excluded phenotype of metastatic colorectal cancer," she suggested, as the dual immune-altering approach "may not be sufficient to generate that immune response that we need for antitumor activity."

The results may be related to the fact that for the majority of patients (90%), tumors were characterized by microsatellite stability, whereas tumors with microsatellite instability have been shown to respond to immunotherapy.

"However, within this trial, extensive biomarker studies were done, and the evaluations, including gene expression analyses, are ongoing and will be presented at a future meeting," she added.

In the postpresentation discussion, Bendell remained upbeat about the potential of the drug combination.

She said: "We know cobimetinib plus atezolizumab was a shot on goal that showed initial promising data, but we can't give up on potentially turning cold tumors to hot tumors, to getting microsatellite stable [disease] to respond to different immunotherapies.

"Various combinations are still in clinical trials," she continued, including "newer immunotherapy agents, different targeted agents in combination with immunotherapies that hopefully will show more promising results and better outcomes in these patients in the future."

However, session cochair Roberto F. Labianca, MD, director of the cancer center at Ospedale Papa Giovanni XXIII, Bergamo, Italy, who was not involved in the study, did not agree with Bendell's assessment.

He told Medscape Medical News: "Of course, I'm disappointed.... I think that this line of approach to CRC patients without microsatellite instability probably is not working. I don't know if other studies in this setting are indicated. Probably not.

"My conclusion is quite pessimistic," he continued, "but the data are negative. It's a very good study, a very large study, an international study and very well designed, so if the results are negative, we have to keep that in mind."

Study Details

Bendell began her presentation by pointing out that for patients with mCRC for whom chemotherapy has failed, the prognosis is poor; the median survival is 6 to 8 months.

Moreover, microsatellite-stable mCRC, which accounts for the vast majority of cases, is considered to be "immune excluded," owing to a lack of tumor-infiltrating lymphocytes, and single-agent immunotherapy has shown minimal activity.

Preclinical evidence has indicated, however, that combining atezolizumab and cobimetinib increases antitumor T-cell responses, and a phase 1b study in heavily pretreated mCRC patients showed that the combination had clinical activity and was well tolerated.

The researchers therefore conducted the phase 3, multicenter, open-label IMblaze370 study in 363 patients with unresectable advanced CRC or mCRC who had received ≥2 prior chemotherapy regimens and whose ECOG performance status was 0-1.

The patients were randomly assigned in a 2:1:1 ratio to receive either atezolizumab 840 mg plus cobimetinib 60 mg (A+C), or atezolizumab 1200 mg alone (A), or regorafenib 160 mg (R) until the loss of clinical benefit.

The primary endpoints were overall survival for A+C vs R and for A vs R. The patients were stratified on the basis of extended RAS mutation status and time since the diagnosis of first metastasis.

The three treatment groups were well balanced. The median age was approximately 58 years, and around 60% of the patients were male.

About a third of these patients (30%) had been diagnosed with a first metastasis less than 18 months earlier, and around 25% of patients had received more than three previous treatment regimens.

Bendell pointed out that approximately 90% of patients had microsatellite-stable disease, and between 34% and 43% of patients whose tumors had been tested for PD-L1 expression showed expression of at least 1%.

The median overall survival was 8.9 months in the A+C arm, 7.1 months in the A arm, and 8.5 months in patients given regorafenib. The proportion of patients alive at 12 months was 38.5%, 27.2%, and 36.6.%, respectively.

There was no significant difference in survival between the groups. For A+C vs R, the hazard ratio was 1.00 (P = .9871); for A vs R, the hazard ratio was 1.19 (P = .3360).

Subgroup analysis revealed few differences in survival outcomes between the treatments, including when stratifying the patients by RAS mutation status and microsatellite instability, although A+C appeared to perform significantly better than R in nonwhite patients, at a hazard ratio for overall survival of 0.40.

There was also no significant difference in progression-free survival between the treatments, at a median of 1.9 months for A+C and C and 2.0 for R. The hazard ratio for A+C vs R was 1.25; that for A vs R was 1.39.

The objective response rates were also almost identical between the treatments, at 2.7% for A+C and 2.2% for A and R. Disease control rates were 26.2%, 21.1%, and 34.4%, respectively.

A+C and R were associated with the highest rates of grade 3-4 adverse events, at 61% and 58%, respectively. For grade 3-4 treatment-related adverse events, the rates were 45% and 49%, respectively. This compares with just 31% and 10%, respectively, for A.

Patients who received A+C were more likely to have an adverse event leading to withdrawal from treatment, at 21% vs 4% for A and 9% for R.

The most common adverse events with A+C were diarrhea (65%), rash (37%), and nausea (37%).

For A, the most common adverse events were fatigue (26%), decreased appetite (24%), and nausea (21%). For R, the most common adverse events were palmar-plantar erythrodysesthesia syndrome (53%), fatigue (46%), and decreased appetite (41%).

The study was sponsored by Hoffmann-La Roche. The study investigators have disclosed no relevant financial relationships.

World Congress on Gastrointestinal Cancer 2018. Abstract LBA-004, presented June 22, 2018.

For more from Medscape Oncology, follow us on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: