STX-3329

Both the number and size of tumours in NF1 patients increase in response to the rise in steroid hormones seen at puberty and during pregnancy. The size of tumours decreases after delivery, suggesting that hormone-targeting therapy might provide a viable new NF1 treatment approach. Our earlier studies demonstrated that human NF1 tumour cell lines either went through apoptosis or ceased growth in the presence of 2-methoxyoestradiol (2ME2), a naturally occurring anticancer metabolite of 17-β estradiol. Previous reports of treatment with sulfamoylated steroidal and non-steroidal derivatives of 2ME2 showed promising reductions in tumour burden in hormone-responsive cancers other than NF1. Here we present the first studies indicating that 2ME2 derivatives could also provide an avenue for treating NF1, for which few treatment options are available.

STX3451, (2-(3-Bromo-4,5-dimethoxybenzyl)-7-methoxy-6-sulfamoyloxy-1,2,3,4-tetrahydroisoquinoline), a non-steroidal sulphamate analogue of 2ME2, was tested in dose-dependent studies of malignant and benign NF1 human tumour cell lines and cell lines with variable controlled neurofibromin expression. The mechanisms of action of STX3451 were also analysed.

We found that STX3451-induced apoptosis in human malignant peripheral nerve sheath tumour (MPNST) cell lines, even in the presence of elevated oestrogen and progesterone. It inhibits both PI3 kinase and mTOR signalling pathways. It disrupts actin- and microtubule-based cytoskeletal structures in cell lines derived from human MPNSTs and in cells derived from benign plexiform neurofibromas. STX3451 selectively kills MPNST-derived cells, but also halts growth of other tumour-derived NF1 cell lines.

STX3451 provides a new approach for inducing cell death and lowering tumour burden in NF1 and other hormone-responsive cancers with limited treatment options.

A SAR translation strategy adopted for the discovery of tetrahydroisoquinolinone (THIQ)‐based steroidomimetic microtubule disruptors has been extended to dihydroisoquinolinone (DHIQ)‐based compounds. A steroid A,B‐ring‐mimicking DHIQ core was connected to methoxyaryl D‐ring mimics through methylene, carbonyl, and sulfonyl linkers, and the resulting compounds were evaluated against two cancer cell lines. The carbonyl‐linked DHIQs in particular exhibit significant in vitro antiproliferative activities (e.g., 6‐hydroxy‐7‐methoxy‐2‐(3,4,5‐trimethoxybenzoyl)‐3,4‐dihydroisoquinolin‐1(2H)‐one (16 g): GI50 51 nM in DU‐145 cells). The broad anticancer activity of DHIQ 16 g was confirmed in the NCI 60‐cell line assay giving a mean activity of 33 nM. Furthermore, 6‐hydroxy‐2‐(3,5‐dimethoxybenzoyl)‐7‐methoxy‐3,4‐dihydroisoquinolin‐1(2H)‐one (16 f) and 16 g and their sulfamate derivatives 17 f and 17 g (2‐(3,5‐dimethoxybenzoyl)‐7‐methoxy‐6‐sulfamoyloxy‐3,4‐dihydroisoquinolin‐1(2H)‐one and 7‐methoxy‐2‐(3,4,5‐trimethoxybenzoyl)‐6‐sulfamoyloxy‐3,4‐dihydroisoquinolin‐1(2H)‐one, respectively) show excellent activity against the polymerization of tubulin, close to that of the clinical combretastatin A‐4, and bind competitively at the colchicine binding site of tubulin. Compounds 16 f and 17 f were also shown to demonstrate in vitro anti‐angiogenic activity. Additionally, X‐ray and computational analyses of 17 f reveal that electrostatic repulsion between the two adjacent carbonyl groups, through conformational biasing, dictates the adoption of a “steroid‐like” conformation that may partially explain the excellent in vitro activities.