I would be extremely surprised to see 50% responses in this patient population, although several patient characteristics could skew the data and result in such a high percentage:
1. Age - older patients tend to have fewer Tregs.
2. Tumor sizes.
3. The number of TIL at baseline for the combo. Remember, the predicted non-responder trial intentionally selected patients who were immunologically frozen, i.e. very few CD8 positive T cells with an exhausted phenotype.
I have not yet observed any combination with an anti-PD-1 that results in anything close to 30% in anti-PD-1 nonresponders in advanced metastatic melanoma, or any solid tumor cancers for that matter.