Correlation between serum ANCA and cancer: a decade-long retrospective analysis of an Italian cohort.

Background: Serum antineutrophil cytoplasmic antibodies (ANCA) are autoantibodies primarily linked with ANCA-associated vasculitides (AAV) but associated diseases include also other autoimmune diseases, infections, and malignancies. However, the association of ANCA with cancer risk remains undefined, especially in patients without AAV. Objective: The aim of this study was to identify the incidence of malignancies in a large cohort during a 10-year period and to compare the incidence in ANCA-positive versus ANCA-negative subjects. Methods: A retrospective cohort study involving 6285 subjects who underwent ANCA testing was performed at Humanitas Research Hospital (ICH) (Rozzano, Italy) from 2007 to 2017, regardless of the clinical indication for ANCA testing. After retrieving comorbidity data from electronic health records, we classified chronic conditions into six major categories and generated a propensity score including age, gender, time of blood draw, and comorbidities. A 1:2 matching was performed, yielding a final cohort of 214 ANCA-positive and 319 ANCA-negative subjects. The ICH Cancer Diagnosis Registry was used to assess cancer incidence. Competing risk regression was performed using the Fine & Gray model to estimate cancer risk associated with pANCA positivity. The cANCA-positive group (n = 44) was excluded from regression analysis due to the absence of cancer events. Results: During a mean follow-up of 3.9 years, 43 patients had a diagnosis of cancer (9/214 ANCA-positive patients, and 34/319 ANCA-negative patients; Chi-square p 0.007). In the pANCA-positive subgroup, no significant difference in cancer incidence was observed compared to ANCA-negative subjects. ANCA-negative subjects did not have significantly longer survival after cancer diagnosis compared with pANCA-positive patients (median survival: 203 days [IQR 120-309] vs. 266 days [IQR 124-580]). In competing risk regression analysis, no association was found between positive pANCA and cancer risk (HR 0.50; CI 95% 0.240-1.043; p 0.065).The results were consistent even when AAV cases were excluded (HR 0.53; 95% CI 0.24-1.16). Conclusion: Isolated ANCA positivity is not associated with a higher incidence of malignancy. There was no significant difference in post-cancer survival between ANCA-negative and pANCA-positive subjects. The absence of malignancies in the cANCA subgroup should be interpreted with caution given the small sample size. These observations should be confirmed by prospective studies, which also should ascertain the possible underlying mechanisms.