麻豆传媒

Patients with inflammatory conditions who have the CC genotype were at greater risk both for discontinuing the use of azathioprine (Imuran) due to hematopoietic toxicity and for needing to switch to lower doses, a retrospective cohort study found.

In a competing-risk analysis involving over 1,400 such patients starting on azathioprine, those who carried the CC genotype of variant rs2814778 – known to be more common among Black individuals – had a nearly three times greater risk of discontinuing azathioprine because of hematopoietic toxicity (HR 2.92, 95% CI 1.57-5.41), compared with TT or TC genotype carriers.

Discontinuation rates were 3.92 per 100 person-years with the CC genotype compared with 1.34 per 100 person-years with the TT or TC genotype.

Those with the CC genotype also had lower last counts of leukocytes (5.8 vs 6.7) and neutrophils (3.1 vs 4.4), as well as lower final azathioprine dose (1.1 vs 1.3 mg/kg per day), reported Cecilia Chung, MD, MPH, of Vanderbilt University Medical Center in Nashville, Tennessee, and colleagues, writing in the .

Discontinuation risk was linked to race alone (HR 2.13, CI 1.21-3.75), but that association lost significance after further adjusting for genotype.

The thiopurine immunosuppressant drug, azathioprine, is for preventing renal transplant rejection and active rheumatoid arthritis. It can be used to treat inflammatory bowel disease (IBD), systemic lupus erythematosus (SLE), and other conditions, Chung's group noted.

However, azathioprine is limited by its narrow therapeutic index and risk of hematopoietic toxicity. The risk for hematopoietic toxicity can more than double in patients prescribed azathioprine who carry certain genetic variants (TPMT or NUDT15), although most patients who discontinue it do not carry those variants.

Benign neutropenia -- lower neutrophil counts without any added risk for infection -- is linked to the ACKR1 variant rs2814778-CC, a single-nucleotide polymorphism (SNP), which is also implicated in malaria resistance.

For this study, Chung and colleagues examined data from the biobank at Vanderbilt University Medical Center (BioVU) encompassing 1,466 patients with inflammatory conditions. Since only the CC genotype for the SNP rs2814778 was linked to lower neutrophil counts, patients were further stratified by having a CC genotype (n=101) or a TT or TC (n=1,365) genotype.

Participants included were new azathioprine users with IBD (18-35%), SLE (10-25%), rheumatoid arthritis, vasculitis, or other connective tissue disorders. The analysis was limited to Black and white patients (165 and 1,301, respectively) with DNA genotype data on file from specialized Illumina technology. Two-thirds of the study population were women. Mean age was 44.

No significant difference between genotype groups was seen in initial dose, year of initial dose, or prior testing for the enzyme that breaks down thiopurines. Not surprisingly, the CC genotype was rare among whites (n=5). Patients with the CC genotype were also more likely to be younger, have SLE, and have lower baseline leukocyte counts.

During a median follow-up of 20.7 months, the risk for azathioprine discontinuation because of hematopoietic toxicity (neutropenia, leukopenia, pancytopenia, thrombocytopenia, or anemia) remained significant even after adjusting for race.

An external cohort of Black children with acute lymphoblastic leukemia (n=94) from the validated the findings on lower dosing among those with the CC genotype who received the thiopurine 6-mercaptopurine compared with the TT or TC genotype (median 0.83 vs 0.94 mg/kg per day).

Chung's group noted that the "normal" cell counts physicians use for decisions on azathioprine discontinuation and dosing are mainly based on values from whites, which might put Black patients at risk for inappropriate dosing or unnecessary discontinuation.

"We recommend that testing for the Duffy-null [CC] phenotype be considered in all patients before azathioprine initiation or if leukopenia is detected while a patient is using azathioprine," they concluded.

The authors acknowledged limitations to the data, such as the inclusion of only two tertiary care centers, and the possibility of unmeasured confounding.

Comment