A new model can predict which patients are more likely to experience side effects from long-term methotrexate (MTX) use, research suggests. Patients with a lower risk profile may benefit from less frequent testing, the authors hypothesize.
Most recommendations advise that patients initiating MTX therapy should get blood testing every 2-4 weeks to monitor for full blood count, liver function, urea electrolytes, and creatinine. After 6 months taking MTX, monitoring can be tapered to every 3 months. But Abhishek Abhishek, MD, PhD, professor of rheumatology and honorary consultant rheumatologist at Nottingham (England) University Hospitals NHS Trust and colleagues argue that abnormal results after the initial 6 months of treatment are “infrequent,” and patients may benefit from fewer tests throughout the year.
“Unnecessary blood tests waste patients’ time and health care resources, including the time of general practitioners and phlebotomists,” Dr. Abhishek and associates write. “It would be beneficial to predict the risk of clinically significant abnormal blood test results during long-term methotrexate treatment to inform the frequency of testing for individuals.”
In the study, published in the BMJ, researchers used the UK’s Clinical Practice Research Datalink (CPRD) to identify the electronic medical records of over 37,000 adult patients with an immune-mediated inflammatory disease who were prescribed MTX during 2007-2019. All included patients were prescribed MTX for at least 6 months. The main outcome was discontinuation of methotrexate because of abnormal blood test results. Around 62% of patients had rheumatoid arthritis and 22% had psoriasis or psoriatic arthritis.
Using these anonymized data, the group developed a risk stratification model using 11 clinical predictors. “The factors that went in the model are simple things that most patients can self-report or doctors can get from their patient’s medical records,” Dr. Abhishek told this news organization, including methotrexate dose, age, sex, and comorbidities. Dr. Abhishek emphasized that the model should be used only in patients who have continued taking MTX for at least 6 months and have already undergone more frequent initial testing.
The strongest individual predictors were diabetes (hazard ratio, 1.25), chronic kidney disease stage 3 (HR, 2.01), and previous cytopenia or raised liver enzyme levels during the first 6 months of MTX therapy (HR, 2.97). However, Dr. Abhishek emphasized that the individual factors were less important, noting that the model sums the risks to predict outcomes more accurately. Most patients (68.4%) were sorted into the low-risk cohort, with a less than 10% estimated risk of discontinuing MTX over the next 5 years. About one-fifth (20.9%) were categorized as moderate risk (10%-20% estimated risk over 5 years), and 10.7% were high risk, with a greater than 20% estimated risk of discontinuing the drug over 5 years.
The authors argue that low-risk patients could receive less frequent testing – perhaps every 6 months or annually, while moderate-risk patients would continue to be tested every 3 months. High-risk patients could potentially be tested with even greater frequently.