If you use a PD-L1-expressing tumor threshold of >5% for treatment with a checkpoint inhibitor monotherapy, then of course you aren't going to observe statistically significant differences. BMY's phase III NSCLC trial design seemed so obviously flawed.
From what I've been reading, there are three things that probably need to happen in order for anti-PD-1/PD-L1 drugs to be effective. First, the tumor must express the PD-L1 protein in relatively high proportion. Apparently, a >50% threshold is working somewhat in NSCLC. If all you're using is a checkpoint inhibitor monotherapy, and the targeted ligand is not proportionally high, then you're only going to eliminate a small proportion of the tumor cells.
Second, you need to have the "substrate" of CD8 "killer" t-cells, i.e. tumor infiltrating lymphocytes (TIL), that will infiltrate the tumors. If you don't have a sufficient number of activated CD8 positive t-cells, then all you will have using anti-PD-1 monotherapy is a relatively small number of TIL. In other words, if you create an adequate immune response by somehow activating t-cells in vivo - or engineer and expand them ex-vivo and inject back into the patient - then you create the "substrate" that is necessary for TIL proliferation.
Third, a significant proportion of the TIL need to be PD-1 positive. From what I have read so far, an approximately >20% bar is necessary to demonstrate a predictable response to anti-PD-1 therapy. If the TIL aren't PD-1 positive to begin with, then there's no point in using anti-PD-1 therapy.
My impression is that both Oncosec and Heat Biologics are able to provide the essential "substrate" - effective numbers of TIL - that would theoretically maximize response rates to anti-PD-L1/PD-1 and other antibody therapies. Both companies appear to be able to achieve significant immune responses with neoantigens, i.e. tumor-matching antigens. In addition, even if the tumor cells don't express PD-L1 ligands or the TIL aren't sufficiently PD-1 positive, then you still end up with a response because you have adequate TIL and memory precursor cells being produced. It makes logical sense for BMS, Merck, et al. to invest in products that will maximize checkpoint inhibitor efficacy.
