Testing blood for a key protein can help identify cancer patients who are likely to benefit from immune checkpoint inhibitors, one of the most promising forms of immunotherapy for cancer, Dana-Farber Cancer Institute researchers report in a new study.
The study, published today in Cancer Immunology Research, found that blood levels of angiopoietin-2 (ANGPT2), a protein involved in blood vessel formation, were a gauge of how well patients with metastatic melanoma responded to checkpoint inhibitors. The study also suggests that ANGPT2 may serve as a target for both immune checkpoint-inhibiting drugs and inhibitors of blood vessel formation (angiogenesis).
“In this study, we found ANGPT2 to be a predictive and prognostic biomarker of response to the inhibitors of immune checkpoints CTLA-4 and PD-1,” said F. Stephen Hodi, MD, director of the Center for Immuno-Oncology at Dana-Farber. “We found additional evidence suggesting a role for angiogenic factors in immune regulation and the possibility of targeting the immune system and angiogeneic factors with combinations in the treatment of cancer.”
Determining which patients are apt to respond to immune checkpoint inhibitors is particularly challenging because of the complexity of cancer cells themselves, as well as the multifaceted nature of the immune system and the tissue surrounding tumors, Hodi explained. A way of cutting through this complexity would be to identify blood biomarkers that are easy to measure and reliably predict checkpoint inhibitor response.
From the outset, ANGPT2 seemed a strong candidate for a biomarker. Not only does it help tumors siphon nutrients from the bloodstream, but it can be a target of antibodies made by patients receiving immune therapies. This suggests that patients who maintain low levels of the protein over the course of treatment might receive particular benefit from immune checkpoint inhibitors.
The study involved three sets of patients with advanced melanoma: 48 who were treated with ipilimumab (Yervoy), which blocks an immune checkpoint protein known as CTLA-4; 43 who were treated with ipilimumab and bevacizumab (Avastin), an antiangiogenic medicine; and 43 who received nivolumab (Opdivo) or pembrolizumab (Keytruda), which targets the PD1 checkpoint protein. All patients had two blood samples taken – one prior to treatment, one less than three months after starting treatment. The patients were from three different clinical trials in which survival data were available.
About 17, 20, and 37 percent of the patients in the three respective studies had their tumors disappear or shrink, and the median follow-up time for all of them was 33 months.
The researchers found that among patients treated with ipilimumab alone, those with high pretreatment levels of ANGPT2 survived a median of 12.2 months, compared to 28.2 months for those with low levels. In those treated with ipilimumab plus bevacizumab, the median overall survival (OS) was 10.9 versus 19.3 months. In patients treated with a PD1 inhibitor, the median OS was 7.3 months among those with high pretreatment ANGPT2; median OS had not yet been reached in those from the low group as more than half of the patients were alive at the time of this report.
To understand how changes in blood ANGPT2 levels with treatment influenced length of survival, the researchers calculated the “fold change” in these levels before and during treatment. (Fold change measures how much a quantity increases or decreases over a particular period of time. A change from 25 to 50, for example, would be a one-fold increase.)
The researchers found that in patients who received ipilimumab alone, those whose ANGPT2 levels changed less and more than 1.25-fold survived a median of 12.4 and 28.1 months, respectively. The same trend held for patients in the ipilimumab plus bevacizumab study, but the difference was not statistically significant.
When the investigators combined the data from all three trials, they found that patients who began with high ANGPT2 levels and experienced large-fold shifts early in the course of treatment had the shortest survival times. Those with low pretreatment levels and small-fold changes, by contrast, had the longest survival. The close relationship between ANGPT2 levels and survival indicate the protein’s potential as a prognostic biomarker.
To understand how ANGPT2 influences survival times, the investigators conducted laboratory tests with a specific subset of immune system cells called macrophages, which often appear in the vicinity of tumors. They found that ANGPT2 produced a thicker mat of the immune checkpoint inhibitor PD-L1 on the cells’ surface.
“These in vitro studies suggested that angiogeneic factors not only play a role in blood vessel formation in tumors but also have immune-regulatory effects, “Hodi said. “Therefore, therapeutics that inhibit angiogenesis could synergize with immune checkpoint blockade. We would, however, like to confirm the findings from this study in a larger patient population in the future,” he added.
The study was in part funded by the National Institutes of Health, the Melanoma Research Alliance, the Sharon Crowley Martin Memorial Fund for Melanoma Research, the Malcolm and Emily Mac Naught Fund for Melanoma Research, Genentech/Roche, Bristol-Myers Squibb, and a Stand Up To Cancer – Cancer Research Institute Cancer Immunology Dream Team Translational Research Grant (SU2C-AACR-DT1012). Hodi has served as a consultant to Genentech and has received research support to his institution from Bristol-Myers Squibb. Patent application for Ang-2 (ANGPT2) as a biomarker is pending.