Primary Resistance to PD-1 Blockade Mediated by JAK1/2 Mutations - PubMed (original) (raw)

doi: 10.1158/2159-8290.CD-16-1223. Epub 2016 Nov 30.

Jesse M Zaretsky 1, Helena Escuin-Ordinas 1, Angel Garcia-Diaz 1, Siwen Hu-Lieskovan 1, Anusha Kalbasi 1, Catherine S Grasso 1, Willy Hugo 1, Salemiz Sandoval 1, Davis Y Torrejon 1, Nicolaos Palaskas 1, Gabriel Abril Rodriguez 1, Giulia Parisi 1, Ariel Azhdam 1, Bartosz Chmielowski 1 2, Grace Cherry 1, Elizabeth Seja 1, Beata Berent-Maoz 1, I Peter Shintaku 1, Dung T Le 3, Drew M Pardoll 3, Luis A Diaz Jr 3, Paul C Tumeh 1, Thomas G Graeber 1 2, Roger S Lo 1 2, Begoña Comin-Anduix 1 2, Antoni Ribas 4 2

Affiliations

Primary Resistance to PD-1 Blockade Mediated by JAK1/2 Mutations

Daniel Sanghoon Shin et al. Cancer Discov. 2017 Feb.

Abstract

Loss-of-function mutations in JAK1/2 can lead to acquired resistance to anti-programmed death protein 1 (PD-1) therapy. We reasoned that they may also be involved in primary resistance to anti-PD-1 therapy. JAK1/2-inactivating mutations were noted in tumor biopsies of 1 of 23 patients with melanoma and in 1 of 16 patients with mismatch repair-deficient colon cancer treated with PD-1 blockade. Both cases had a high mutational load but did not respond to anti-PD-1 therapy. Two out of 48 human melanoma cell lines had JAK1/2 mutations, which led to a lack of PD-L1 expression upon interferon gamma exposure mediated by an inability to signal through the interferon gamma receptor pathway. JAK1/2 loss-of-function alterations in The Cancer Genome Atlas confer adverse outcomes in patients. We propose that JAK1/2 loss-of-function mutations are a genetic mechanism of lack of reactive PD-L1 expression and response to interferon gamma, leading to primary resistance to PD-1 blockade therapy.

Significance: A key functional result from somatic JAK1/2 mutations in a cancer cell is the inability to respond to interferon gamma by expressing PD-L1 and many other interferon-stimulated genes. These mutations result in a genetic mechanism for the absence of reactive PD-L1 expression, and patients harboring such tumors would be unlikely to respond to PD-1 blockade therapy. Cancer Discov; 7(2); 188-201. ©2016 AACR.See related commentary by Marabelle et al., p. 128This article is highlighted in the In This Issue feature, p. 115.

©2016 American Association for Cancer Research.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Mutational load and mutations in interferon signaling pathway among patients with advanced melanoma with or without response to anti-PD-1 blockade therapy

A) Total non-synonymous mutations per tumor from biopsies of patients with response (n=14) or without response (n=9) to anti-PD-1 per RECIST 1.1 criteria (median 503 versus 274, P = 0.27 by Mann-Whitney). Median and interquartile range are shown, with value for each individual tumor shown as dots. B–D) Each column corresponds to an individual case from panel A. B) Depiction of mutational load (bar graph) and mutations in interferon receptor pathway genes. The size of circles and adjacent labels represent the tumor variant allele frequency (VAF) after adjustment for stromal content. Color represents predicted functional effect. Green = missense; orange = nonsense. Red circle highlights amplified JAK1 mutation in one patient who did not respond to anti-PD-1 therapy. All the tumor sequences were compared to normal germline sequences. C) Heat map of the density of CD8 T cells in the invasive margin or intra-tumor compartment analyzed in baseline tumor biopsies by immunohistochemistry. D) Heat map of density of PD-L1 expression in available tissue samples. E) Genetic amplification of the chr9p24.1 (PD-L1, PD-L2 and JAK2 locus, termed the PDJ amplicon) was noted in one biopsy from a non-responding patient. Heat map represents average read depth ratio vs paired germline normal.

Figure 2

Figure 2. Altered interferon signaling with JAK1 loss of function mutation in M431 and interferon gamma-inducible PD-L1 expression by 48 melanoma cell lines

A) Mean fluorescent intensity of PD-L1 expression by flow cytometry upon interferon alpha, beta or gamma exposure over 18 hours in M431 (established from patient #15) compared with M438 (established from patient #8). B) Corresponding Western blot analyses for M431 upon interferon exposure for 30 minutes or 18 hours. C) Phosphorylated STAT1 (pSTAT1) flow cytometry for M431 upon interferon exposure for 30 minutes or 18 hours (same color scale was in Fig. 2). The numbers in the heat map of pSTAT1 indicate the average Arcsinh ratio from two independent phospho-flow cytometry experiments. D) PD-L1 response to interferon gamma. Blue arrows represent average change from baseline upon interferon gamma exposure. Grey shades show the full range of measured values (n=2 or 3). Red stars indicate cell lines with no response and black stars indicate cell lines with poor response to interferons. Red: BRAF mutated; blue: NRAS mutated; green: BRAF and NRAS mutated; black: BRAF wild type, NRAS wild type.

Figure 3

Figure 3. Defects in interferon receptor signaling pathway with JAK homozygous loss of function mutations in M368 and M395

A and B) Exome sequencing data showing JAK2 D313 spice site mutation in exon 8 in M368 (A), and JAK1 D775N kinase domain mutation in exon 17 in M395 (B). Upper plot shows individual sequencing reads using the Integrated Genomics Viewer (IGV); lower plot shows position relative to kinase domains using the cBioPortal Mutation Mapper. C and D) For each cell line, cells were cultured with interferon alpha, interferon beta or interferon gamma for either 30 minutes or 18 hours, or with vehicle control (c, first column from the left in Western blots and phospho-flow data). Phosphorylated STAT1 (pSTAT1) detected by Western blotting (top panel) or phospho-flow cytometry data (bottom panel). The numbers in the heat map of pSTAT1 indicate the average Arcsinh ratio from two independent phospho-flow experiments. Blots represent two independent replicate experiments. E and F) PD-L1 expression after interferon exposure on M395 and M431 after JAK1 wild-type lentiviral transduction respectively. G and H) Time course PD-L1 expression for M431 and JAK1 wild type lentiviral vector transduced M431 respectively.

Figure 4

Figure 4. Mutational burden of somatic, protein-altering mutations per subject from whole exome sequencing for patients with advanced colon cancer who participated PD-1 blockade clinical trial

A) Similar to Fig. 1B, bar graph shows mutational load in individual cases (fraction SNVs, blue; insertions, red; deletions, orange) divided by response to PD-1 blockade therapy. Bottom panel depicts mutations, insertions, or deletions in the interferon receptor pathway. Color represents predicted functional effect. The size of circles and adjacent labels correspond to tumor variant allele frequency (VAF) after adjusting for stromal content. Red circle highlights homozygous nonsense mutation in JAK1 from one of patient who did not respond to anti-PD-1 therapy. B) Sequencing reads of JAK1 mutation in non-responder subject # 12. C) Mutation observed in 51 reads out of 80, (VAF 0.64) which corresponds to a homozygous mutation (adjusted VAF 0.94) when adjusted for a tumor purity of 68%. D) Copy number profile reveals loss of heterozygosity across most of the genome, including chromosome 1/JAK1.

Figure 5

Figure 5. Analysis of JAK1 and JAK2 mutations in the Cancer Cell Line Encyclopedia database

A) Variant allele frequency (left axis, red and blue points) and percentage of tumors with mutations in JAK1 or JAK2 (right axis, grey bars) in CCLE database from the cBioPortal. B) Non-synonymous mutational burden was analyzed for individual cell lines (each dot represents cell line) and plotted for each histological type. JAK1 or JAK2 mutated cell lines were color coded (red: VAF>0.75 and blue: VAF<0.75).

Figure 6

Figure 6. Frequency of JAK1 and JAK2 alterations and their association with overall survival in TCGA datasets

Kaplan-Meier survival analysis of TCGA skin cutaneous melanoma (A), breast invasive carcinoma (B) and prostate adenocarcinoma (C) provisional datasets, comparing control patients (blue) and patients harboring specified alterations in JAK1 and JAK2 (red). Frequency and distribution of combined JAK1 and JAK2 alterations are shown within each set of Kaplan-Meier plots. Significance testing of overall survival was performed using log-rank analysis.

Comment in

Similar articles

Cited by

References

    1. Herbst RS, Soria JC, Kowanetz M, Fine GD, Hamid O, Gordon MS, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;515:563–7. - PMC - PubMed
    1. Powles T, Eder JP, Fine GD, Braiteh FS, Loriot Y, Cruz C, et al. MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer. Nature. 2014;515:558–62. - PubMed
    1. Robert C, Long GV, Brady B, Dutriaux C, Maio M, Mortier L, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med. 2015;372:320–30. - PubMed
    1. Ansell SM, Lesokhin AM, Borrello I, Halwani A, Scott EC, Gutierrez M, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin’s lymphoma. N Engl J Med. 2015;372:311–9. - PMC - PubMed
    1. Robert C, Schachter J, Long GV, Arance A, Grob JJ, Mortier L, et al. Pembrolizumab versus Ipilimumab in Advanced Melanoma. N Engl J Med. 2015;372:2521–32. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources