Identification of factors determining patterns of serum CRP level reduction in response to anti-tuberculosis treatment initiation

Introduction. In patients undergoing lengthy anti-tuberculosis (TB) therapy, the monitoring of treatment response is the cornerstone of effective disease management. The serum C-reactive protein (CRP) levels are reported to complement the clinical, radiological, and bacteriological findings at the time of diagnosis, but failure to achieve a reduction in the levels of this biomarker by half within the first two weeks of anti-TB treatment was associated with adverse treatment outcomes.
Objectives. This study aimed to clarify the role of the patient baseline characteristics and plasma exposure of four first-line anti-TB drugs (rifampicin, isoniazid, pyrazinamide, ethambutol) in the early reduction of the inflammation in response to anti-TB treatment.
Methods. The study enrolled patients with drug-susceptible pulmonary TB (n = 41) admitted to the Riga East University Hospital, Centre of Tuberculosis and Lung Diseases. All the clinically relevant data, including serum CRP levels at baseline and 10–12 days after treatment onset, were extracted from the electronic medical records. The anti-TB drug plasma concentration was measured in three time points (pre-dose, 2 h and 6 h post-dose) using a validated LC-MS/MS method. The data analysis was conducted using IBM SPSS Statistics 29.
Results. Following 10–12 days of anti-TB treatment, the median serum CRP level in the study population decreased from 22.8 to 6.4 mg/L (W = 142, p = 0.001). Nearly one third of patients (31.7%) had serum CRP levels within the reference range at both time points (Group A). In 34.1% of patients, serum CRP levels decreased ≥2 times from the baseline (Group B), while the same proportion of patients did not meet this criterion (Group C). Overall, the baseline characteristics were consistent across all three groups, except for radiological and bacteriological findings. Patients in Group A less frequently presented with lung cavitations compared to those in Group B (adjusted p = 0.003) and Group C (adjusted p <0.001), as well as positive sputum-smear microscopy result at the baseline compared to Group C (adjusted p = 0.012). On the contrary, the extent of lung involvement (unilateral vs. bilateral damage) did not differ among the groups (p = 0.909). The assessment of anti-TB drug exposure revealed that median rifampicin and isoniazid plasma concentration 2 h post-dose (Cmax) was below the therapeutic range (1.8 vs. 8.0 µg/mL and 2.4 vs. 3.0 µg/mL, respectively), whereas for remaining drugs, it was within the target range. Comparing the anti-TB drug exposure among the three groups, the estimated marginal means of Cmax and area under the time-concentration curve (AUC0–6h) slightly differed but failed to reach significance.
Conclusion. The results of this study suggest that disease severity may delay the decrease in inflammation associated with anti-TB treatment. Despite the observed variability, anti-TB drug plasma exposure did not directly impact changes in serum CRP levels resulting from treatment initiation. The clinical relevance of these findings is yet to be investigated.