Objective:

The article examines the objective, findings, and clinical context described in Reply to Zhou and Shan: Understanding long-term rituximab responses in immune thrombocytopenia.

Approach:

• Cohort inclusion: Patients required at least two registered ITP diagnoses more than 6 months apart; medical-record validation included confirmation of thrombocytopenia.
• Corticosteroid co-administration: The authors assessed its association with sustained response among patients not previously treated, while noting incomplete dose data and limited ability to separate treatment effects.
• Sustained-response analyses: All 503 initial responders were retained in the 2- and 5-year analyses, including patients without sufficient follow-up to reach those endpoints.
• Target-trial feasibility: The authors explain that cohort eligibility, treatment initiation, and follow-up could not be aligned, and that treatment strategies and serial platelet data were not consistently available.

Key Findings:

• Routine specialist follow-up for at least 12 months after rituximab in the Nordic countries was described as mitigating concern that durable early responders were underrepresented.
• Patients whose disease course ended within 6 months could not be included; any resulting bias would underestimate response duration, but its effect on the reported median of 5.4 years is unknown.
• Among non-pretreated patients, corticosteroid co-administration was associated with sustained response at 5 years (risk ratio 8.29; 95% CI 2.35–29.16); the authors cautioned that the wide interval reflects uncertainty and does not establish causality.
• Using all 503 initial responders, sustained response proportions were 53.3% at 2 years and 31.4% at 5 years. Among patients with adequate follow-up, the corresponding proportions were 54.7% (268/490) and 38.4% (158/411).
• Complete response was described as an early signal of sustained response, not a pretreatment criterion for selecting patients for rituximab.

Interpretation:

The authors characterize the sustained-response estimates using all initial responders as conservative because some patients lacked adequate follow-up. They state that routine clinical characteristics largely do not predict initial response and that complete response may help identify patients likely to remain in remission.

Limitations:

• Patients completing their disease course within 6 months of initial contact were not eligible; the impact on the median response duration cannot be determined.
• Only 32 of 123 previously untreated patients received rituximab without corticosteroids, and corticosteroid-dose information was missing or incomplete for many patients.
• The study end date limited follow-up for some patients; 13 had less than 2 years of follow-up and 79 had at least 2 but less than 5 years.
• The registry could not support target-trial emulation because eligibility could be confirmed only after treatment for many patients, upfront rituximab was predominantly used in Denmark, and serial platelet counts were not systematically available.

Conclusion:

The authors acknowledge that the registry design limits causal comparisons and response-duration estimates, and clarify that complete response is relevant to follow-up assessment rather than pretreatment selection.

Sources:

• Article context supplied in the prompt