Virtual Library

Start Your Search

Chih-Yuan Hsu



Author of

  • +

    P2.04 - Immuno-oncology (ID 167)

    • Event: WCLC 2019
    • Type: Poster Viewing in the Exhibit Hall
    • Track: Immuno-oncology
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall
    • +

      P2.04-50 - Advanced Statistical Approach Tells the Difference: Taylor-Expansion Adjustment for Survival Analyses in Immunotherapy Trials (ID 1781)

      10:15 - 18:15  |  Presenting Author(s): Chih-Yuan Hsu

      • Abstract
      • Slides

      Background

      The success in immune checkpoint inhibitors (ICIs) has again revolutionized the paradigm in lung cancer therapy, especially among patients without druggable driver mutations. Landmark studies, such as Keynote 010/042 and Checkmate 017/057, showed that patients in the ICI study arm had a better overall survival compared to patients receiving standard chemotherapy. Furthermore, most of the Kaplan-Meier (KM) survival curves crossed within the first 3 to 6 months and the survival curve for ICI study arm showed a long-tail, suggesting the existence of a subgroup that truly respond to ICIs. However, the traditional Cox proportional hazards model does not provide adjustment for treatment responses between true-responders and poor-responders. We hypothesize that hazard ratios (HRs) between the treatments differ from these two subgroups and adjustment is required for survival evaluation in ICI trials.

      Method

      Overall survivals in Keynote 010, 042, 189, 407 and Checkmate 017/057 three-year updated data served as the real data illustrations in the current study. The Taylor expansion was applied to the adjustment of HRs derived from the traditional Cox model. Comparisons between the adjusted and the crude HRs were made and the differences in the proportion of patients with long-term survival obtained by the adjustment approach were also inferred.

      Result

      Our results showed that when the adjustment approach was applied, the adjusted HRs for the ICI poor-responders compared to the chemotherapy control became statistically non-significant. Furthermore, the proportion of patients with long-term survival differed significantly between ICI true-responders and chemotherapy control. The pattern was observed in all four ICI monotherapy studies with crossing survival curves in the initial 3-6 months (Table 1).

      slide1.jpg

      Conclusion

      The results suggested that the traditional Cox model might not be the optimal method for survival analyses in ICI trials and adjustment is necessary. The differences between the adjusted and the crude HRs may be very significant.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.