Dr. Eric Whitman's Very Positive Quick Summary of DCTH Presentation
A phase III random assignment trial comparing percutaneous hepatic perfusion with melphalan (PHP-mel) to standard of care for patients with hepatic metastases from metastatic ocular or cutaneous melanoma.
Well, this presentation went very well, I thought. The data was even better than the intial press release from April. The hepatic PFS (primary endpoint) was 245 days in the PHP arm and 49 days in the BAC (best alternative care) arm. There was no survival benefit on an intent to treat basis BUT (and this point is currently rattling the cyber sphere) 55% of the 49 BAC patients crossed over to PHP after progressing. Since OS analysis is based on MEDIAN survival of the group, if over 50% of BAC patients end up eventually getting PHP AND benefit similarly from the delayed PHP, then the survival curves should superimpose, as they do. The only way in a crossover trial for the OS curves to NOT superimpose when over 50% crossover would be if the delay in getting PHP resulted in a lesser clinical effect (which clearly was not the case). There was an interesting survival curve they showed comparing the BAC patients divided into whether or not they received PHP at any point. The curves were WIDELY separated, but the analysis was not prespecified and carries no analytic weight statistically. Overall PFS also signficant although less so than hepatic PFS, which is logical. This was 189 days vs 46 days. The response rate data was: 34.1% PR, 52.3% SD. It was noted that NONE of the patients who actually received PHP progressed on initial assessment. There is no question in my mind that this is clinically effective and I was surprised when the discussant, seemingly from left field, pooh-poohed the results and said it was biased, ineffective and not worthy??? The format of the presentations prevented any questioning or commenting of the discussants.
