1. Cumulative exposure to biologics and risk of cancer in psoriasis patients: a meta-analysis of Psonet studies from Israel, Italy, Spain, United Kingdom and Republic of Ireland.
I. Garcia-Doval, M.A. Descalzo, K.J. Mason, A.D. Cohen, A.D. Ormerod, F.J. Gómez-García, S. Cazzaniga, I. Feldhamer, H. Ali, E. Herrera-Acosta, C.E.M. Griffiths, R. Stern, L. Naldi, on behalf of the Psonet Network.
Fundación Academia Española de Dermatología y Venereología, Madrid, Spain
British Journal of Dermatology. DOI: /bjd.16715

2. Ignacio Garcia-Doval, MD, MSc Epid, PhD

3. Introduction What’s already known?
• Systemic therapies, including biologic therapies, are widely prescribed for patients with moderate-to-severe psoriasis • Due to the immunomodulatory mechanism of biologic therapies for psoriasis, it is hypothesized that long-term exposure to them may increase the risk of developing cancer

4. Methods
Objective
To describe if, in patients with psoriasis requiring systemic therapy, there was a dose-response effect linking cumulative exposure to biologics with a higher risk of cancer.

5. Methods Four Psonet studies provided data for this study:
British Association of Dermatologists Biologic Interventions Register (BADBIR)
Spanish Registry of Systemic Therapy in Psoriasis (Biobadaderm)
Clalit Health Service Database
Italian Registry of Systemic Therapy in Psoriasis (Psocare)

6. Methods
BADBIR, Biobadaderm and Psocare are prospective pharmacovigilance registries of psoriasis patients receiving biologic and/or systemic therapies
Clalit Health Service database comprises electronic medical records including prescriptions provided by the main health service provider in Israel

7. Methods Nested case-control studies were performed in each registry
Outcome definition
All incident cancers; prevalent cancers, and benign and in-situ tumours were excluded
Lists of cancers from each registry reviewed by two clinicians (IGD and LN) for inclusion

8. Methods Cases: patients with incident cancers
Controls: up to four cancer-free patients in each cohort were matched to each case by year of birth, sex and registration hospital (all registries), and year of entry into registry (BADBIR, Clalit, Psocare)

9. Analysis
Cumulative exposure to biologic therapies, methotrexate and ciclosporin calculated by totalling the person years of follow-up on each therapy from initiation to stop or censor (date of cancer diagnosis for cases and their matched controls)

10. Analysis
Three conditional logistic regression models with increasing adjustment (adj.) produced by each registry to investigate risk of cancer
Crude: cumulative exposure to biologic therapies
Adj. 1: crude model & cumulative exposure to methotrexate and
ciclosporin
Adj. 2: model 2 & duration of psoriasis (years from diagnosis to
censor); smoking status; prevalence or length of exposure to
phototherapy; and the modified Charlson Comorbidity Index
(CCI)

11. Analysis Sensitivity analyses
Main dataset (MD): risk of cancer, excluding basal cell carcinoma (BCC) & squamous
cell carcinoma (SCC)
MD+SCC: risk of cancer including SCC, but excluding BCC
MD+SCC+BCC: risk of cancer, including BCC and SCC
Results from each registry combined in prospective meta-analysis using random-effects model

12. Results
Total study population: 27,376 patients with 58,978 person-years of follow-up
728 cases who developed a cancer matched with 2,671 controls
41 cases of SCC and 108 cases of BCC in BADBIR, Biobadaderm and Psocare
SCC and BCC data not available from Clalit

13. Results
After matching, differences between cases and controls were present for the CCI, and prevalence of previous psoralen- ultraviolet-A and smoking (BADBIR only)

14. Forest plot: meta-analysis of risk of cancer in Psonet registries
Random effects meta-analysis
Crude: odds ratio of cancer risk (skin cancer excluded) per year of cumulative lifetime exposure to biologics
Adj. 1: odds ratio of cancer risk with additional adjustment for cumulative exposure to methotrexate and ciclosporin
Adj. 2: odds ratio of cancer risk with additional adjustment for duration of psoriasis, smoking, exposure to phototherapy and modified Charlson Comorbidity Index
OR per year of exposure

15. Forest plot: meta-analysis of sensitivity analyses
Estimates of the risk of cancer per year of cumulative lifetime exposure to biologics
Sensitivity analyses: main dataset (MD); MD+SCC; MD+SCC+BCC
Adj. 1: odds ratio of cancer risk with additional adjustment for cumulative exposure to methotrexate and ciclosporin
Adj. 2: odds ratio of cancer risk with additional adjustment for duration of psoriasis, smoking, exposure to phototherapy and modified Charlson Comorbidity Index
OR per year of exposure

16. Results
The risk of first cancer was not significantly associated with cumulative exposure to biologics
Adjusted odds ratio per year of exposure 1.02; (95%CI 0.92, 1.13)
Results were similar if SCC and BCC were included in the outcome

17. Discussion What does this study add?
This study suggests that treatment of psoriasis patients with biologics in current clinical practice is not associated with an increased risk of cancer in the medium-term, after a few years of use and latency
Given the heterogeneity of the data sources and methodological limitations, it remains unclear if use of biologics may be associated with an increased (albeit small) risk of certain cancer subtypes.

18. Call for correspondence
Why not join the debate on this article through our correspondence section?
Rapid responses should not exceed 350 words, four references and one figure
Further details can be found here