Background Before the availability of mRNA vaccines, many transplant centers chose to significantly reduce maintenance immunosuppression in kidney transplant recipients (KTRs) with SARS‐CoV‐2 infection. The extent to which this increases the risk of allosensitization is unclear. Methods In this observational cohort study, we analyzed 47 KTRs from March 2020 to February 2021 who underwent substantial reduction of maintenance immunosuppression during SARS‐CoV‐2 infection. KTRs were followed at 6 and 18 months concerning the development of de novo donor‐specific anti‐HLA (human leukocyte antigen) antibodies (DSA). The HLA‐derived epitope mismatches were calculated using the predicted indirectly recognizable HLA‐epitopes (PIRCHE‐II) algorithm. Results In total, 14 of 47 KTRs (30%) developed de novo HLA antibodies after the reduction of maintenance immunosuppression. KTRs with higher total PIRCHE‐II scores and higher PIRCHE‐II scores for the HLA‐DR locus were more likely to develop de novo HLA antibodies (p = .023, p = .009). Furthermore, 4 of the 47 KTRs (9%) developed de novo DSA after reduction of maintenance immunosuppression, which were exclusively directed against HLA‐class II antigens and also showed higher PIRCHE‐II scores for HLA‐class II. The cumulative mean fluorescence intensity of 40 KTRs with preexisting anti‐HLA antibodies and 13 KTRs with preexisting DSA at the time of SARS‐CoV‐2 infection remained stable after the reduction of maintenance immunosuppression (p = .141; p = .529). Conclusions Our data show that the HLA‐derived epitope mismatch load between donor and recipient influences the risk of de novo DSA development when immunosuppression is temporarily reduced. Our data further suggest that reduction in immunosuppression should be made more cautiously in KTRs with high PIRCHE‐II scores for HLA‐class II antigens.