Germline TYK2 mutation and cancer risk

Background

On September 1, 2021 the U.S. Food and Drug Administration (FDA) announced that three Janus kinase (JAK) inhibitors approved to treat chronic inflammatory conditions would require safety warnings over increased rates of serious heart-related events, lung cancer, and lymphoma¹. Recently, oral Deucravacitinib, a selective inhibitor of Tyrosine kinase 2 (TYK2, a member of the JAK family), was shown to lead to larger improvements in symptom severity for patients with moderate-to-severe plaque psoriasis than oral Apremilast, the current standard of care^2,3. Deucravacitinib, therefore, has the potential to become an important treatment option for patients with psoriasis requiring systemic treatment and is currently under review for approval at the FDA and European Medicines Agency². It is unclear, however, whether the elevated cancer risk associated with some JAK inhibitors is shared with this novel TYK2 inhibitor.

In the absence of long-term clinical trial data, naturally occurring genetic variation can be leveraged to validate therapeutic targets and predict their adverse effects⁴. Specifically, germline mutations causing partial or complete loss-of-function (LOF) of genes encoding drug targets can be employed to mimic pharmacological inhibition of these targets and have been used to correctly predict adverse effects of new medications⁵.

Here, we used an established partial LOF mutation in TYK2 (rs34536443), previously shown to protect against psoriasis and other autoimmune diseases^6,7, to evaluate the potential effect of therapeutic TYK2 inhibition on risk of lung cancer and non-Hodgkin lymphoma.

Methods

Minor allele homozygosity of rs34536443 causes near-complete (∼80%) loss of TYK2 function, while heterozygotes have a more modest reduction in function (<40%), suggesting non-additive effects of this variant⁶. To validate this variant as a surrogate for therapeutic TYK2 inhibition, we evaluated the effect of each copy of the minor allele of rs34536443, representing partial TYK2 inhibition, on risk of psoriasis, inflammatory bowel disease, Crohn’s disease, and multiple sclerosis. These analyses were performed using summary genetic association data on up to 78,334 cases and 150,030 controls from genome-wide association studies (GWAS) of these autoimmune diseases^8-11.

To evaluate the effect of this variant on risk of overall and histological subtype-specific lung cancer we obtained summary genetic association data on up to 29,266 cases and 56,450 controls from a GWAS meta-analysis of the Integrative Analysis of Lung Cancer Risk and Etiology (INTEGRAL)¹². Summary genetic association data for non-Hodgkin lymphoma (NHL) were generated by meta-analysing genome-wide association (GWAS) data for rs34536443 from the UK Biobank and InterLymph consortium in METAL¹³.

In secondary analyses, we explored whether rs34536443 increased risk of three other common adult cancers (breast, colorectal, prostate), which, along with lung cancer, account for approximately half of all new U.S. cancer cases¹⁴. Summary genetic association data on overall and histological subtype-specific breast, colorectal, and prostate cancer on up to 260,346 cases and 234,774 controls were obtained from analyses of the Breast Cancer Association Consortium (BCAC), Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO), ColoRectal Cancer Transdisciplinary Study (CORECT), Colon Cancer Family Registry (CCFR), and the Prostate Cancer Association Group to Investigate Cancer Associated Alterations in the Genome (PRACTICAL) consortium^15-17.

All analyses were restricted to participants of European ancestry. Further information on statistical analysis, imputation, and quality control measures for these studies is provided in Supplementary Material and in the original publications. All studies contributing data to these analyses had the relevant institutional review board approval from each country, in accordance with the Declaration of Helsinki, and all participants provided informed consent.

Results

To validate rs34536443 as a surrogate for therapeutic TYK2 inhibition, we confirmed that each copy of the minor allele, representing partial TYK2 inhibition, was associated with lower risk of psoriasis (OR 0.48, 95% CI 0.39-0.58, P = 9.01 × 10^-14) and other autoimmune diseases (Table 1).

In analyses of 29,266 lung cancer cases and 56,540 controls, each copy of the minor allele of rs34536443 was associated with an increased risk of lung cancer (OR 1.15, 95% CI 1.07-1.24, P = 1.72 × 10^-4). This association was stronger for current and former smokers (OR 1.15, 95% CI 1.09-1.23, P = 2.29 × 10^-6) compared to never smokers (OR 1.09, 95% CI 0.91-1.32, P = 0.34). The magnitude of effect was similar across histology types. In analyses of 8,489 cases and 374,506 controls, each copy of the minor allele of rs34536443 was associated with an increased risk of non-Hodgkin lymphoma (OR 1.18, 95% CI 1.05-1.33, P = 5.25 × 10^-3). In secondary analyses, rs34536443 was weakly associated with risk of advanced prostate cancer, but not associated with other cancers examined.

Discussion

The efficacy of Deucravactinib in treating plaque psoriasis is attributed to the selective inhibition of TYK2, a downstream mediator of pro-inflammatory signalling pathways critical to psoriasis³. Using an established partial loss-of-function variant to mimic therapeutic TYK2 inhibition, we show that potential protection from autoimmunity mediated by TYK2 inhibition may be counteracted by an increased risk of lung cancer and non-Hodgkin lymphoma. This finding, including the restriction of an increased lung cancer risk to current and former smokers, is consistent with recent reports of higher rates of these two cancers in users of the JAK inhibitor Tofacitinib in the ORAL Surveillance safety trial¹. Importantly, the per-allele estimates presented in this analysis may underestimate the effect of therapeutic TYK2 inhibition on cancer given larger TYK2 reductions achieved by Deucravacitinib at doses shown to confer therapeutic benefit (50-80% TYK2 inhibition) compared with the <40% expected per copy of the rs34536443 minor allele¹⁸. These findings, suggesting potential adverse target-mediated effects of TYK2 inhibition on lung cancer and non-Hodgkin lymphoma, could have important implications for future safety assessment of Deucravacitinib and other TYK2 inhibitors in development.

Funding

JY is supported by a Cancer Research UK Population Research Postdoctoral Fellowship (C68933/A28534). JY, RMM, and GDS are supported by a Cancer Research UK (C18281/A29019) programme grant (the Integrative Cancer Epidemiology Programme). JY, KB, RMM, and GDS are part of the Medical Research Council Integrative Epidemiology Unit at the University of Bristol which is supported by the Medical Research Council MC_UU_00011/1, MC_UU_00011/5) and the University of Bristol. RMM is also supported by the NIHR Bristol Biomedical Research Centre which is funded by the NIHR and is a partnership between University Hospitals Bristol NHS Foundation Trust and the University of Bristol. The US National Cancer Institute supports JM and PB (UO1CA203654) and JM (UO1CA257679). Department of Health and Social Care disclaimer: The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the International Agency for Research on Cancer or the Department of Health and Social Care.