Long noncoding RNA signatures may help predict immunotherapy response
medwireNews: Chinese researchers have used a combination of long noncoding RNA (lncRNA) signatures and cytotoxic T-lymphocyte (CTL) tumor infiltration to identify four distinct immune classes that may predict response to cancer immunotherapy.
Erwei Song and colleagues from Sun Yat-sen University in Guangzhou, China, explain that lncRNAs “are involved in innate and adaptive immunity in cancer by mediating the functional state of immunologic cells, pathways, and genes,” but their role in cancer immunotherapy “is largely unknown.”
They therefore analyzed lncRNA and genomic data from 348 patients with bladder cancer treated with the PD-L1 inhibitor atezolizumab as part of the IMvigor210 trial and 71 patients with melanoma from The Cancer Genome Atlas (TCGA) who received various immunotherapy agents, including PD-1 and CTLA-4 inhibitors, and a cytokine tumor vaccine.
They identified 49 lncRNAs that could be used to create two distinct lncRNA-based classes based on overall survival (OS). The researchers designated these two groups as immune-functional and immune-nonfunctional due to differences in immune cell expression between them. Individuals in the immune-functional class had a significant 35% lower risk for death at 2 years than those in the immune-nonfunctional class.
Additional analysis among 2951 patients with various cancers from TCGA who were immunotherapy-naïve suggested that CTL infiltration may reflect the majority of immunologic variance in the tumor microenvironment.
With this in mind, Song and team further classified the patients with bladder cancer into four groups: an immune-active class that included patients with an immune-functional lncRNA signature and a high density of CTLs; an immune-exclusion class (immune-functional lncRNA signature plus low CTL density); an immune-dysfunctional class (immune-nonfunctional lncRNA signature plus high CTL density); and an immune-desert class (immune-nonfunctional lncRNA signature plus low CTL density).
Median OS differed significantly among these classes, at not reached versus 9.6, 8.1, and 6.7 months, respectively, which the researchers say suggests “a close interaction between dysregulated lncRNA biology and the dysfunctional state of immune cells.”
Song et al also used 29 of the 49 lncRNAs to create an lncRNA score. Patients with bladder cancer who had a low lncRNA score had a significant 68% lower risk for death than those with a high score.
Moreover, when this score was added to the combination of tumor mutational burden, PD-L1 expression, and CTL infiltration, the combined algorithm strongly correlated with survival at both 12 and 20 months, with areas under the receiver operating characteristic curve (AUC) of 0.81 and 0.80, respectively. By comparison, the AUC in a model that did not include the lncRNA score was 0.63.
Writing in JAMA Network Open, Song and co-authors say that “[t]he 4 novel distinct classes identified in this study indicated that immune molecular classification of aspects of both immune exclusion and immune dysfunction could be informative for understanding patterns of immune escape (ie, cancers could escape immunologic destruction) and for selection of candidates for cancer immunotherapy.”
The conclude by recommending that “factors characterizing immune cell dysfunction, such as the lncRNA score, should be integrated into multiomic panels for precision immunotherapy.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2020 Springer Healthcare part of the Springer Nature Group