“Long-term routine laboratory toxicity monitoring (lt-RLTM) is used for management of patients with rheumatoid arthritis (RA) who are being treated with disease-modifying antirheumatic drugs (DMARDs). Intensive monitoring is recommended during the first 6 months due to higher risks for adverse reactions, followed by less frequent monitoring for the duration of DMARD use, which can last for decades. Laboratory parameters used for lt-RLTM typically include liver, renal, and hematologic toxicity markers. Despite high consistency in the types of tests ordered, the frequency of testing and responses to abnormal results vary widely across practices.
[..] Typically, lt-RLTM is performed every 3 months for conventional synthetic DMARDs (csDMARDs) for monotherapy and combination therapies and for rituximab and tocilizumab monotherapy. Other biologic DMARDs (bDMARDs) and hydroxychloroquine are mostly judged not to require lt-RLTM. The necessity of such frequent lt-RLTM for patients on stable DMARD doses is debated, particularly because reduced monitoring during the COVID-19 pandemic did not cause evident harm. Limiting lt-RLTM practices to what adds value is important, as reducing lt-RLTM frequency could lessen patient and health care burden and costs. [..]
Methods
This retrospective cohort study was performed at the rheumatology department of the Sint Maartenskliniek, a specialized rheumatology practice in the Netherlands with a main location in Nijmegen and outpatient clinics within 7 other general hospitals. [..]
All patients with a clinical diagnosis of RA in outpatient care between July 2008 and April 2020 who had used a DMARD for at least 6 months (index event) were considered for inclusion. [..]
Routine laboratory toxicity monitoring was conducted in all patients at weeks 4, 8, 12, and 24 after the start of DMARD treatment, followed by lt-RLTM every 3 to 6 months. Additional monitoring was recommended during dose escalation and as indicated based on a physician’s judgment. Overall, care adhered closely to current RA treatment recommendations from the European Alliance of Associations for Rheumatology. [..]
The test set within this cohort included 5 laboratory parameters, with results categorized as abnormal and very abnormal: alanine aminotransferase (ALT; >100 and >300 U/L), estimated glomerular filtration rate (eGFR; <60 and <45 mL/min/1.73 m2), hemoglobin (Hb; <7.5 mmol/L [females] or <8 mmol/L [males] and <6 mmol/L), leukocyte count (<3.5 and <2.0 × 109/L), and platelet count (<140 and <100 × 109/L). Cutoffs for abnormal and very abnormal results were based on guidelines, derived from several clinical treatment recommendations on when clinical intervention is deemed necessary, or based on expert opinion. A single abnormal or very abnormal result was counted as an event. [..]
The primary outcome is the cumulative incidence of new abnormal or very abnormal lt-RLTM results at 2 and 5 years after initiation of individual or combination DMARD therapy, calculated overall and by drug type or combination. In addition, the absolute number of abnormal and very abnormal results is reported per laboratory parameter and DMARD type. [..]
Results
Of the 5341 patients with RA who had long-term use of DMARDs, 4774 met inclusion criteria regarding the DMARD combinations used, with total follow-up of 18 383 patient-years. The median time between treatment initiation and the first lt-RLTM test set was 9.4 months (IQR, 7.1 to 31.4 months). [..]
Among the 297 775 lt-RLTM tests, 17 343 (5.8%) results were abnormal and 2064 (0.69%) results were very abnormal, the latter occurring in 492 unique patients, of which 449 were new abnormal results. [..]
Although the proportion of abnormal results across all laboratory tests was relatively low, the cumulative probabilities of abnormal results for the 5 tests at the patient level varied from 4.4% to 47% at 2 years and from 7.5% to 61% at 5 years after left- and right-censoring were accounted for. The cumulative probabilities of very abnormal results for the 5 tests varied between 0.2% and 6.6% at 2 years and between 0.3% and 11% at 5 years.
Discussion
Our results show that very abnormal lt-RLTM results are uncommon in patients using long-term DMARD treatment, with cumulative 5-year incidences ranging from 0.4% for leukocyte count less than 2.0 × 109/L to 11% for eGFR less than 45 mL/min/1.73 m2. In addition, in-depth review of new very abnormal results revealed that they were often already known, occurred after a dose increase, were not related to DMARD use, or resulted in no action. The incidence of less serious abnormal results was higher, as high as 39% for eGFR below 60 mL/min/1.73 m2 and 61% for Hb level below 7.5 mmol/L for females or below 8 mmol/L for males.
The low incidence of very abnormal lt-RLTM results found in this relatively large study suggests the general safety of DMARDs in patients with RA and aligns with data from large cohorts and clinical trials on the safety of these drugs, which show that long-term DMARD treatment rarely causes new clinically relevant hematologic, liver, or renal adverse effects. It is also important to note that even when adverse events occur, prevention through lt-RLTM is generally not possible. The adverse events that occur are often linked to the disease itself; for example, RA is associated with increased risk for lymphoma and cardiovascular disease. In addition, even adverse events that may be causally related to the drug are often not targetable for prevention by laboratory monitoring. For example, the small excess infection risk for tocilizumab is generally not mediated by leukopenia, and the slightly higher risk for nonmelanoma skin cancer for methotrexate also has no measurable substrate in blood. Overall, our results fit with existing epidemiologic and pathophysiologic data and considerations, although they may seem counterintuitive for some physicians and patients.
[..] Laboratory testing is generally more frequent in the first 6 months of treatment, and as we do not challenge the value of this, this initial phase was beyond the scope of our research question. By starting follow-up at this later point, we avoided immortal time bias and focused on estimating risk during the more stable, long-term phase of DMARD use rather than during the initial months, when laboratory abnormalities are more common and monitoring is more intensive. For some patients, the first included laboratory test occurred well beyond 6 months after treatment initiation; in fact, the 75th percentile was at 31 months. Although left-censoring was addressed analytically, this delayed entry may have led to underestimation of the true duration of drug exposure. Consequently, the cumulative incidence of abnormalities might be slightly overestimated, as the number of events is divided by a shorter observation period. The infrequent occurrence of abnormal results, even under these conservative assumptions, strengthens the conclusion that clinically significant abnormalities during long-term DMARD use are uncommon. Data were also derived from practice settings where long-term laboratory monitoring was generally protocolized clinically in all patients, regardless of received DMARD treatment.
[..] In the absence of definitive data, we believe that a reasonable strategy could be to conduct routine laboratory monitoring in the first 6 months of DMARD use and less frequent monitoring thereafter (for example, every 1 to 2 years), with closer monitoring when the drug dosage is increased, to evaluate suggestive signs or symptoms, and in patients at high risk for toxicity or with recent abnormal laboratory results.
In conclusion, the long-standing practice of performing frequent lt-RLTM in all patients with RA using DMARDs should be reconsidered, as our findings call into question the value of nontargeted monitoring. More selective laboratory testing would reduce patient burden, environmental toll, and health care costs.”
Full article, E Ulijn, N Broeder, K Bevers, et al. Annals of Internal Medicine, 2025.8.26