A good post Jase99. I'd add few things on different topics...

A good post Jase99. I'd add few things on different topics discussed here lately.

The referenced T-VEC vs GM-CSF trial recruited during 2009-2011, and at the time there was not many treatment options aside from surgical resection. For 50% of pts T-VEC/GM-CSF was the FIRST systemic therapy (Table 1) and half of the CRs were in Stage IIIb category (Fig 2A), which implies spread to lymph nodes but no metastatic lesions seen (e.g. google melanoma pathological staging).

In VAXINIA, the minimum requirement for recruitment is 2 or more previous systemic therapies. As we live in 2024, the recruited 3 melanoma pts (2/3 PR and 1 unknown) have had previous ICIs such as pembro for sure. In general, to recruit patients in Ph1 studies they should have had all the available approved therapies for his or her tumor type at the time of the recruitment. Therefore CRs are quite unique in Ph1 studies. In 465 NCI sponsored solid tumor Ph1 trials (N=13,800), doi: 10.1016/S0140-6736(22)01390-3, there were 2.7% CRs overall. In combination arms 3.7% CRs vs monotherapy arms 0.3% CRs (supplement Table8). Other reviews of Ph1 solid tumor trials has found roughly same kind of results from what I've seen. In the 2010s CRs have become more prevalent in Ph1, but the investigated drugs have become more specific.

In 2022 results from T-VEC+pembro vs pembro RCT in melanoma were reported, doi: 10.1200/JCO.22.00343. For 346+346 pts, CRs were 18% vs 12%. Durable response 42% vs 34%. The results imply that there isn't much synergy between T-VEC+pembro, but both work for immunologically 'warm/hot' tumors. Compared to CF33-hNIS that in vitro and in murine models has shown to increase PD-L1 expression on tumor cells and be synergistic with PD-L1 drug, doi: 10.1080/2162402X.2020.1729300.

On the topic of dosage, I'm amazed how many think that from these results that we have seen, it could somehow be inferred what the optimal dose is. Tumors and their anticipated (e.g. melanoma vs pancreatic) + observed responses for immunotherapy vary widely as can be seen from the waterfall plot in the poster. All we can observe is different patients with one particular dose. From this setting, it requires quite lot of imagination to infer what would have happened to the exactly same patient if the dose would have been 1e6, 1e7 or 1e8. The most recent poster also shows that the lymphocyte counts and other measurable factors are all over the place; the within-group variance is larger than the between-group variance with this small sample. As it is known that melanoma, lung, kidney, bladder cancers are much more likely to response to the currently available immunotherapies than some other tumor types (pancreatic, bile duct, sarcoma etc), the observed differences in lymphocyte reactions likely represent as much the unique responsiveness of the tumor to CF33 than the dose given.

Lastly on the expansion cohorts. As I read the FDA manual / guidance, in the expansion cohort itself, it is possible to dose-escalate if there is scientific rationale for it and it is mentioned in the expansion cohort protocol (FDA-2018-D-2777, page 14).

Cheers from Finland