This article is more than 5 years old. Information may no longer be current.
Tool predicts immunotherapy response among patients with melanoma
Click Here to Manage Email Alerts
A novel gene expression predictor effectively predicted whether patients with melanoma would respond to immune checkpoint inhibitors, according to study results published in Nature Medicine.
Noam Auslander, PhD, of the Center for Cancer Research at NCI and the Center for Bioinformatics and Computational Biology at University of Maryland, developed the predictor — known as IMmuno-PREdictive Score (IMPRES) — with other researchers from those institutions, as well as Harvard University and University of Pennsylvania.
In an analysis that included nearly 300 samples from several studies, IMPRES outperformed existing predictors by identifying nearly all patients with melanoma who responded to immune checkpoint inhibitors and more than half of those who did not.
Researchers hope to evaluate its use in other cancer types.
HemOnc Today spoke with Auslander about the need for a tool to predict immunotherapy response in melanoma, the reliability of IMPRES, and the potential benefits it could offer if it becomes widely used.
Question: Can you describe the need for a tool to predict immunotherapy response in melanoma?
Answer: Immunotherapy — and specifically immune checkpoint blockage therapy — shows remarkable clinical gains in advanced melanoma. However, not all patients respond, and the treatments are associated with adverse effects and costs. Hence, a tool that may help identify patients who are unlikely to benefit from these treatments could be highly valuable.
Q: How did you identify this predictor?
A: IMPRES is a simple score counting immune relations (such as A>B) between 15 pairs of genes. The list of relations was composed to predict tumor spontaneous regression in neuroblastoma, because we hypothesized that the immune mechanisms underlining spontaneous regression are similar to those underlining the response to immune checkpoint blockage therapy.
Q: How reliable is it?
A: IMPRES robustly predicted the response to immune checkpoint blockage therapy for almost 300 melanoma samples, from pre- and early-on treatment samples, treated with different therapies (both anti PD-1 and anti CTLA-4).
Q: Why hasn’t something like this been done before?
A: There is a shortage of large cohorts of immune checkpoint blockade-treated patients that include both ‘omics’ and response data, and the data sets we used to test IMPRES were not available until quite recently. Second, patient immune response is complex. Using binary relations via IMPRES enabled us to construct an accurate predictor that maintains its performance across data sets and platforms, which was not possible previously for this task.
Q: What must still be confirmed in future research?
A: The results of our early study are encouraging, but a clinical trial is necessary to prove the clinical utility of IMPRES. Further research is needed to confirm efficacy in additional patient data sets — particularly in other cancer types for which checkpoint inhibitors have been approved.
Q: What other studies are planned or underway?
A: We plan to run a clinical trial to test the utility of IMPRES. When more data becomes available, we will evaluate the performance of IMPRES in other tumor types.
Q: What are the potential benefits if this becomes widely used?
A: IMPRES may spare patients who are unlikely to respond to immune checkpoint blockage therapy from the associated adverse effects and costs. Additionally, it may potentially help decide which patients should be treated with combination of drugs and which are likely to respond to a single drug.
Q: Is there anything else that you would like to mention?
A: Our results also demonstrate some similarities between advanced melanoma and neuroblastoma with respect to their immune-related transcriptome. This may indicate shared immune mechanisms underlying the spontaneous regression in these two cancer types and may support the possible potential utility of immune checkpoint blockage therapy in neuroblastoma. – by Jennifer Southall
Reference:
Auslander N, et al. Nat Med. 2018;doi:10.1038/s41591-018-0157-9.
For more information:
Noam Auslander, PhD, can be reached at noam.auslander@nih.gov.
Disclosure: Auslander reports no relevant financial disclosures.