You are correct to question this study. Even in the announcement it stated, “comparisons with other studies have limitations which need to be considered”. IMO you bring up some good points and some not so good points. The truth is we will not know the real survival benefit of Veyonda until NOX runs a large randomised, double-blind clinical trial against LuPSMA alone, so you are correct to question the 71% survival benefit. In saying that, the next best thing is the Vision study, followed potentially by the WARMTH study.
In your previous post you mentioned the “patients were older, had higher PSA, more lung and liver metastases and median of 3 cycles vs 5 cycles LuPSMA”. Based on the 2020 publication (I don’t have access to 2021), mOS was the same regardless of age (<70 vs >70) and there was no statistically significant survival disadvantage for those who received Ra-223 (p = 0.354) or had lung metastases (p = 0.743). There was however a highly significant survival disadvantage for those who had liver metastasis (p < 0.0001), with a mOS of 6.045 (4.744–7.346) vs 12.977 (11.306–14.649). But as you can see even if you strip out this group, Veyonda still confers a survival advantage using mOS of 19.7 months.
You also mentioned these patients could have a worse prognosis? One way to compare this is to use the ECOG score. For the WARMTH study, over 80% of the patients have an ECOG score of 0 or 1, while less than 20% have an ECOG score of 2, and median baseline PSA was 177 ng/ml (consistent with the LuPIN trial), so I don’t think these patients necessarily had a worse prognosis. There was also no significant difference in survival for men with a Gleason score or <7 vs >7 (p = 0.132).
Now you may have made a good point by bringing up the median number of cycles of LuPSMA (3 vs 5), although it is difficult to determine exactly why only a median of 3 cycles may have been given. Some considerations need to be given to the fact this study was retrospective and it gathered data from 2014-2018. Something else that has not been mentioned in this study was dosage.
Since these trials involved testing a radioisotope, earlier trials needed to find the minimally effective dose; that is start off with a lower dosage and reduced number of cycles and build a safety database. Since this safety database has now been built, the treatment cycles now follow a set paradigm; that is, 7.4 GBq administered by i.v. infusion every 6 weeks for a maximum of 6 cycles. So it is possible the reduced number of cycles for the WARMTH study was by design and that these patients also received a lower dose, although this is speculation on my part.
There is also another consideration that needs to be made. Since LuPSMA is only given to patients who show a PSA response, the subset of patients who demonstrated a poor PSA response appears to have skewed the overall survival data to the left (see KM curve below). However, for patients who did show a PSA response mOS was found to be around 14 months, still less than the 19.7 months observed for the LuPIN trial.
As mentioned earlier, results from the Vision study will be more comparable to LuPIN than the WARMTH study, so we will find out if Veyonda is de-risked further next Monday. There is no doubt still quite a bit of risk.
(20min delay)