Description

Overcoming Resistance to PD-1 Blockade Therapy

Antoni Ribas M.D., Ph.D.

Department of Medicine, Division of Hematology/Oncology; Department of Surgery; Department of Medical and Molecular Pharmacology, University of California Los Angeles (UCLA), and the Jonsson Comprehensive Cancer Center at UCLA, Los Angeles, CA

Primary resistance to PD-1 blockade therapy is most frequently mediated by a lack of intratumoral tumor antigen-specific T cell infiltration, as if there are no T cells they cannot be inhibited by PD-1:PD-L1 interactions inside the tumor. This could be because the patient’s immune system has not been able to recognize the cancer cells differentially from normal cells, which could be corrected by tumor antigen vaccine approaches. It is possible that antitumor T cells have been primed to the tumor antigens, but that these T cells are inhibited upstream of the cancer such as when CTLA-4 blocks T cell activation in the lymph nodes; in this case, a combination of anti-CTLA-4 and anti-PD-1 therapy could induce tumor responses. Other immune checkpoints may have similar effects and releasing them could synergize with PD-1 blockade therapy. Alternatively, not enough immune activation signaling could be limiting an antitumor T cell response, and combining therapy with immune activating antibodies to CD137/41BB, OX40 or CD40 may provide this signal. There may be immune suppressive factors in the tumor that need to be blocked, such as TGFb, IDO or arginase, or immune suppressive cells such as Treg or myeloid suppressor cells, and several pharmacological strategies are being developed to inhibit these factors and cells.

Finally, if a cancer had a genetic alteration that led to inability to respond to interferons, then this cancer would not be protecting itself from immune attack by expressing PD-L1, and therefore it would be useless to block it. We identified genetic mutations in JAK1 and JAK2 controlling signalling downstream of the interferon receptor that prevent PD-L1 upregulation on melanoma cells upon interferon gamma exposure, which are associated with primary resistance to PD-1 blockade in patients. Whole exome sequencing of metastatic melanoma biopsies from 23 patients treated with pembrolizumab revealed a JAK1 missense mutation with chromosomal amplification in the tumour with the highest mutational load in a non-responding patient. In a second cohort of 16 patients with mismatch repair deficient colon cancer treated with PD-1 blockade, a homozygous JAK1 loss-of-function mutation was similarly found in the tumour of a patient without an objective response to PD-1 blockade. In this situation, it is unlikely that a combination immunotherapy strategy will be able to reverse primary resistance to PD-1 blockade therapy.

Acquired resistance to PD-1 blockade therapy develops in approximately 25-30% of patients who initially had an objective tumor response to therapy. Comparing biopsies of melanoma before starting on anti-PD-1 therapy with the tumors that relapsed months to years later, we again found new JAK1/2 loss of function mutations, and truncating mutations in the antigen presentation gene beta-2 microglobulin (B2M). JAK1 and JAK2 truncating mutations resulted in a lack of response to interferon gamma, including insensitivity to its anti-proliferative effects on cancer cells. The B2M truncating mutation led to loss of surface expression of the major histocompatibility complex (MHC) class I. Recognizing these pathways to acquired resistance suggests that a better front line therapy that eliminates all residual cancer cells using upfront combination immunotherapy would be more beneficial than trying to reverse acquired resistance once it has been established.