The article contains several points requiring clarification. Most importantly, Figure 4 states that the red diagnostic intervals have significantly worse survival than both the favorable PABC intervals and non-PABC controls, although Table 1 provides only descriptive five-year survival probabilities and no subgroup-specific hazard ratios, confidence intervals, or P-values. The authors also state that adjusted interval-specific hazard ratios could not be estimated and that no multiplicity correction was applied. We therefore ask the authors to provide the complete statistical output and exact comparisons underlying Figure 4, including event counts, log-rank results, hazard ratios, confidence intervals, handling of matching, and multiplicity.
In addition, the Methods report 2,735 initially screened patients, whereas Figure 1 reports 2,755; the article reports outcome data for 753 PABC patients, whereas Table 1 indicates 35 missing survival outcomes among 787 patients, leaving 752; and the reported combination of R 4.3.1 with the base stats package 4.4.2 appears inconsistent. Clarification of these discrepancies and correction of the affected text or figures would allow readers to determine whether they are reporting errors or affect the subgroup conclusions.