that covered approximately 24,000 patients and 86,000 person-years of follow-up.
The findings could “help guide clinical decision making,” providing “reassurance that it remains safe to use conventional immunomodulators, anti-TNF [tumor necrosis factor] agents, or newer biologics in individuals with [immune-mediated diseases] with a prior malignancy consistent with recent guidelines,” Akshita Gupta, MD, of Massachusetts General Hospital, Boston, and coinvestigators wrote in Clinical Gastroenterology and Hepatology.
And because a stratification of studies by the timing of immunosuppression therapy initiation found no increased risk when treatment was started within 5 years of a cancer diagnosis compared to later on, the meta-analysis could “potentially reduce the time to initiation of immunosuppressive treatment,” the authors wrote, noting a continued need for individualized decision-making.
Ustekinumab, a monoclonal antibody targeting interleukin-12 and IL-23, and vedolizumab, a monoclonal antibody that binds to alpha4beta7 integrin, were covered in the meta-analysis, but investigators found no studies on the use of upadacitinib or other Janus kinase (JAK) inhibitors, or the use of S1P modulators, in patients with prior malignancies.
The analysis included 31 observational studies, 17 of which involved patients with inflammatory bowel disease (IBD). (Of the other studies, 14 involved patients with rheumatoid arthritis, 2 covered psoriasis, and 1 covered ankylosing spondylitis.)
Similar levels of risk
The incidence rate of new or recurrent cancers among individuals not receiving any immunosuppressive therapy for IBD or other immune-mediated diseases after an index cancer was 35 per 1,000 patient-years (95% confidence interval, 27-43 per 1,000 patient-years; 1,627 incident cancers among 12,238 patients, 43,765 patient-years), and the rate among anti-TNF users was similar at 32 per 1,000 patient-years (95% CI, 25-38 per 1,000 patient-years; 571 cancers among 3,939 patients, 17,772 patient-years).
Among patients on conventional immunomodulator therapy (thiopurines, methotrexate), the incidence rate was numerically higher at 46 per 1,000 patient-years (95% CI, 31-61; 1,104 incident cancers among 5,930 patients; 17,018 patient-years), but was not statistically different from anti-TNF (P = .92) or no immunosuppression (P = .98).
Patients on combination immunosuppression also had numerically higher rates of new or recurrent cancers at 56 per 1,000 patient-years (95% CI, 31-81; 179 incident cancers, 2,659 patient-years), but these rates were not statistically different from immunomodulator use alone (P = .19), anti-TNF alone (P = .06) or no immunosuppressive therapy (P = .14).
Patients on ustekinumab and vedolizumab similarly had numerically lower rates of cancer recurrence, compared with other treatment groups: 21 per 1,000 patient-years (95% CI, 0-44; 5 cancers among 41 patients, 213 patient-years) and 16 per 1,000 patient-years (95% CI, 5-26; 37 cancers among 281 patients, 1,951 patient-years). However, the difference was statistically significant only for vedolizumab (P = .03 vs. immunomodulators and P = .04 vs. anti-TNF agents).
Subgroup analyses for new primary cancers, recurrence of a prior cancer, and type of index cancer (skin cancer vs. other cancers) similarly found no statistically significant differences between treatment arms. Results were similar in patients with IBD and RA.