Comparative efficacy and safety of treatment regimens for Pneumocystis jirovecii pneumonia in people living with HIV - A systematic review and network meta-analysis of randomized controlled trials

Background: Pneumocystis jirovecii pneumonia (PCP) is a serious opportunistic infection in people living with HIV (PWH) who have low CD4 counts. Despite its side effects, trimethoprim-sulfamethoxazole (TMP-SMX) is currently considered the primary treatment for PCP.

Objectives: To compare the efficacy (treatment-failure and mortality) and tolerability (treatment change) of PCP treatment-regimens with a frequentist network meta-analysis (NMA).

Data sources: Embase, MEDLINE, and CENTRAL from inception to February 3, 2024.

Study eligibility criteria: Comparative randomized controlled trials (RCTs) of at least two PCP treatment regimens.

Participants: PWH.

Interventions: Treatment regimens for PCP compared head-to-head.

Assessment of risk-of-bias: Cochrane risk-of-bias tool for RCTs (RoB2) DATA SYNTHESIS: Title, abstract, and full-text screening, along with data extraction, were conducted by two independent reviewers. Data on PCP treatment-failure, all-cause mortality, and discontinuation due to toxicity were pooled and ranked.

Results: Fourteen RCTs conducted between 1983 and 1996 included 1,788 participants across 27 treatment-arms. No regimen showed statistically significant superiority over TMP-SMX in direct comparison. In the network meta-analysis, Clindamycin/Primaquine ranked best (surface-under-the-cumulative-ranking-curve, 0.8), followed by IV Pentamidine (0.8) and TMP-SMX (0.8) regarding treatment-failure. Regarding all-cause-mortality, TMP-SMX was superior to atovaquone in direct comparison, but no treatment was superior in the full network analysis. Dapsone-TMP (0.7) and IV Pentamidine (0.8) ranked highest for mortality reduction. For safety and tolerability, comparator drugs consistently outperformed TMP-SMX, with significant reductions in toxicity observed for Dapsone-TMP, inhaled Pentamidine, and Atovaquone. Inhaled Pentamidine (0.9) was the best tolerated, followed by Trimetrexate (0.8) and Atovaquone (0.8).

Conclusions: We conclude that TMP-SMX should be reassessed as the stand-alone first-line therapy for PCP in PWH, given the better tolerability and comparable efficacy of other treatments. In places with access to alternative drugs for PCP treatment, our analysis suggests that alternative regimens may offer comparable effectiveness, providing flexibility to use alternative treatments when comorbidities necessitate it.