Abstract

Background: Hypomethylating agents and histone deacetylase inhibitors are licensed for myelodysplastic syndrome and cutaneous T-cell lymphoma and remain absent from standard treatment of solid tumours despite two decades of trials. The disparity is usually described as a translational failure awaiting a solution. Methods: Reviews and clinical trials of DNA methyltransferase or histone deacetylase inhibitors in malignancy reporting a response outcome were eligible. Reported response rates were compared across compartments, and the relationship between mechanism and proliferative fraction was set out. Trial-level endpoints were compared within a single solid tumour study reporting response, disease control and duration. No pooling was performed. Analyses were performed in Python 3. Results: In haematological malignancy, vorinostat produced partial responses in approximately 30% of patients with advanced cutaneous T-cell lymphoma, and a class I selective inhibitor produced tumour reduction in 63% of patients with T-cell, follicular and Hodgkin lymphoma including after allogeneic transplantation. In solid tumours, azacitidine with entinostat produced objective responses in 4% of 45 heavily pretreated patients with non-small cell lung cancer, and histone deacetylase inhibitors showed negligible activity in colorectal cancer as monotherapy or in combination. The gap spans 7.5- to 15.8-fold on these figures. Hypomethylating agents are nucleoside analogues incorporated during DNA synthesis and therefore act only on cells traversing S phase, so the fraction of a tumour the mechanism can reach is bounded by its growth fraction—which differs between compartments by roughly the same order as the response gap. In the same lung cancer trial, disease control at 12 weeks reached at least 27% against an objective response rate of 4%, one complete response lasted 14 months and one partial response 8 months, responses continued after the epigenetic agents were discontinued, and five patients subsequently received checkpoint blockade. Conclusions: The haematological-solid dichotomy is a prediction of the mechanism rather than an unexplained translational failure: an agent requiring DNA replication cannot reach the non-cycling majority of a solid tumour. The pattern in the lung cancer trial—low response, substantially higher disease control, durability outlasting exposure and subsequent checkpoint sequencing—is more consistent with a priming effect than with cytotoxicity. If that is what these agents do in solid tumours, response rate measured during exposure is not the endpoint that would detect it.

Keywords: epigenetic therapy; hypomethylating agents; histone deacetylase inhibitors; growth fraction; solid tumours; endpoint selection