Survival benefits may exceed expectations, global applicability awaits validation
AI summary card
Survival benefits may exceed expectations, global applicability awaits validation
Morgan Stanley pointed out that the HARMONi-6 study shows Ivonescimab + chemotherapy significantly extends survival compared to Tislelizumab + chemotherapy, with a median overall survival of 27.9 months versus 23.7 months, but the global applicability requires attention to issues such as data maturity and population bias.
- A median overall survival difference of 4.2 months, but the actual effect size (HR=0.66) may be underestimated
- Survival curves begin to stably separate at around 6 months, with gaps widening at 12/18/24 months
- Discussants focus on data maturity, control group survival rate, population bias, and other global applicability issues
- Researchers believe that data maturity and male bias reflect smoking population characteristics in China, with age distribution similar to previous studies
Report interpretation
Overview
Morgan Stanley interprets key points from the HARMONi-6 study presented at the ASCO 2026 conference, concluding that the study shows Ivonescimab combined with chemotherapy significantly outperforms Tislelizumab combined with chemotherapy in terms of overall survival, with potential survival benefits exceeding surface data, but global applicability still needs verification.
Core views
The report indicates that in the HARMONi-6 study, the Ivonescimab + chemotherapy group achieved statistical significance (HR=0.66) compared to the Tislelizumab + chemotherapy group, with median overall survival times of 27.9 months and 23.7 months respectively. Notably, the 4.2-month median survival difference may not fully reflect the actual effect size because only one event did not reach unrestricted (NR). From about 6 months onwards, the survival curve stably separates, with survival gaps continuing to widen at 12/18/24 months. Although discussants expressed caution regarding global applicability (insufficient data maturity, higher survival rates in the control group than benchmarks, male/elderly/non-PD-L1 expressing population bias), researchers believe that data maturity and male bias reflect smoking population characteristics in China, with age distributions similar to previous studies (such as KN407), helping alleviate these concerns.
Analysis framework
Analysts interpret clinical trial data by focusing on the association mechanisms between overall survival (OS) and progression-free survival (PFS). They first confirm the statistical significance of the study data, then analyze key factors such as data maturity and population bias raised by discussants concerning global applicability concerns, using historical data and Chinese population characteristics to argue for the reliability of the study results.
Methodology notes
Survival analysis and hazard ratio (HR) interpretation
In biotech clinical trials, the hazard ratio (HR) is a core indicator of treatment efficacy. When HR<1, it indicates a higher survival rate in the treatment group; in this study, HR=0.66 indicates a significant increase in survival rates in the treatment group, but should be interpreted comprehensively along with the median survival difference and follow-up time.
Key data
- Hazard Ratio (HR)0.66Statistical significance (95% confidence interval 0.50-0.87, one-sided P=0.0017)
- Median Overall Survival27.9 months (Ivonescimab group)Control group 23.7 months, only 1 event short of unrestricted
- Follow-up TimeApproximately 21 months204 events, limited data maturity
Impact & implications
The report believes that the study results are positive for the biotechnology industry, supporting the mechanism from progression-free survival (PFS) to overall survival (OS), which helps resolve controversies over survival durability. Data from the global HARMONi-3 study will be a key validation point, and if the data is good, it will provide strong support for the global promotion of related drugs.
Risks
- Insufficient data maturity (only 204 events, approximately 21 months of follow-up)
- Higher survival rate in the control group than benchmark levels
- Bias in males/elderly/non-PD-L1 expressing populations
What to watch
- Maturity of the tail end of overall survival
- Data from the global HARMONi-3 study
- VEGF toxicity and proteinuria conditions
- Age and PD-L1 subgroup analyses