Command Palette

Search for a command to run...

Preprint WatchModerateOctober 10th, 2026

PIK-75 suppresses an SREBP-centred sterol programme in PIK3CA-altered small-cell lung cancer cells, but attribution to p110α remains unresolved

Movaqar, A.; Yaghoubi, A.; Rezaee, S. A. R.

PIK-75 suppresses an SREBP-centred sterol and lipid synthesis programme in PIK3CA-altered small-cell lung cancer cells, with attribution of the effect to p110-alpha explicitly unresolved

Moderate contradiction

2 prior failures

Two documented clinical failures match this mechanism, or a single Phase 3 failure is on record.

This re-analysis reports that PIK-75 suppresses an SREBP-controlled sterol and lipid synthesis programme in PIK3CA-altered small-cell lung cancer cells, with log2 fold-changes of -1.1 to -1.6 for SQLE, HMGCS1, HMGCR, LDLR and SCD, while stating that attribution of the effect to p110-alpha remains unresolved. Claidex holds two PIK3CA failures, alpelisib-olaparib-pik3ca-ovarian-epik-o-phase3-efficacy-failure and alpelisib-epik-b3-tnbc, the first an efficacy failure and the second an enrollment collapse. The unresolved isoform attribution is the point of contact: a program that reads a transcriptional response as evidence of on-target p110-alpha inhibition inherits exactly the ambiguity this preprint declines to resolve. Isoform-selective claims in PIK3CA-altered tumours need genetic confirmation rather than pharmacological inference.

Abstract excerpt

Abstract: Background. Class IA PI3K isoforms are candidate therapeutic targets in small-cell lung cancer (SCLC). The transcriptional consequences of isoform-directed inhibitors in SCLC cells have not been compared genome-wide. Methods. We re-analysed GEO series GSE40564: PIK3CA-altered NCI-H69 cells treated for 24 h with vehicle, PIK-75 (p110-directed) or TGX-221 (p110{beta}-directed), three arrays per group (Affymetrix HG-U133 Plus 2.0). Differential expression used limma, with re-processing from raw files and spike-in-anchored normalisation as sensitivity analyses. Gene sets were assessed by effect size, competitive and rotation tests, and size-matched random-set nulls. Isoform dependency was examined in DepMap CRISPR screens. A PIK-75 response score was tested in 81 primary SCLCs against random-signature nulls. Results. PIK-75 reduced SREBP-controlled sterol- and lipid-synthesis genes (SQLE, HMGCS1, HMGCR, LDLR, SCD; log2 fold-change -1.1 to -1.6). Cholesterol, mTORC1, hypoxia and glycolysis gene sets fell far more than size-matched random sets. E2F-target and homologous-repair genes fell more modestly, including BRCA1 (-0.91). The response was broad: 11,645 of 17,458 genes changed at FDR < 0.05, but only 316 decreased and 20 increased significantly beyond two-fold, an asymmetry that persisted under spike-in-anchored normalisation. TGX-221 changed 358 genes, none two-fold. Its set-level effects were four- to nine-fold smaller and depended on two of three replicates. PIK-75 did not induce canonical FOXO targets, and it inhibits kinases other than p110, so the response cannot be attributed to p110 alone. In DepMap, SCLC lines depended more on PIK3CA than on PIK3CB (median Chronos -0.38 vs -0.15), as did other lineages. In patients, the response score was not associated with survival or subtype beyond what random gene signatures achieve (empirical p = 0.14 and 0.11). Conclusion. PIK-75, but not TGX-221, suppresses an SREBP-centred anabolic transcriptional programme in NCI-H69 cells. Attributing this to p110 requires confirmation with selective inhibitors or genetic perturbation.

Matching Claidex post-mortems

2 of 2 indexed

This is an automated contradiction flag, not an editorial judgment on the preprint's quality. Flags identify where the preclinical literature and the clinical failure record diverge.