From:  Synergistic attack of “old drugs for new uses”: repositioning strategies in cancer combination therapy

 Evidence hierarchy of repurposed non-oncology drugs in cancer combination therapy.

Evidence levelRepresentative drug(s)Cancer type(s)Key finding(s)Refs.
I. Confirmatory clinical (Phase III/IV)--No non-oncology drug has yet achieved confirmatory Phase III evidence as a combination partner.-
II. Promising early-phase Clinical (Phase I/II)Metformin (+ chemo)Ovarian, Breast, LymphomaReduced cancer stem cells; enhanced chemosensitivity; improved survival in triple-positive breast cancer.[122124]
II. Promising early-phase clinical (Phase I/II)Valproic acid (+ simvastatin/chemo)Pancreatic ductal adenocarcinomaVESPA trial ongoing; epigenetic “priming” to overcome chemoresistance.[115]
III. Translational preclinical (In Vivo/Organoid)Sertraline (+ artemether)Breast cancer (Ser/Gly-addicted)Potent synergy by targeting serine/glycine synthesis addiction; clear mechanistic basis.[75]
III. Translational preclinical (In Vivo/Organoid)Auranofin (+ AKT inhibitor)Pancreatic cancerPDTO models identified CA12 as a biomarker; personalized repurposing approach.[99]
III. Translational preclinical (In Vivo/Organoid)Pirfenidone (+ chemo)VariousRemodels CAF-rich TME; improves drug penetration; novel delivery systems in development.[113, 120]
IV. Observational/EpidemiologicalStatins, MetforminVariousEpidemiological associations with improved cancer outcomes; generated repurposing hypotheses.[8, 99]
V. Inconclusive/Failed TranslationChloroquine/HydroxychloroquineGlioblastoma, Multiple MyelomaPreclinical synergy failed to translate; lack of pharmacodynamic biomarkers was a major shortcoming.[69]

Interpretation of the Evidence Hierarchy: Several important observations emerge from this classification. First, a significant translational gap exists: despite hundreds of preclinical studies demonstrating synergy, no repurposed non-oncology drug has yet achieved Phase III confirmatory evidence as a combination partner in cancer therapy. Second, the most successful signals (e.g., metformin, valproic acid) share common features: they are mechanistically well-understood, target fundamental hallmarks of cancer (metabolism, epigenetics), and are being evaluated in biomarker-enriched or mechanistically driven trial designs. Third, the CQ experience serves as a cautionary tale, illustrating that robust preclinical autophagy inhibition does not guarantee clinical success without rigorous pharmacodynamic monitoring and patient selection. Collectively, these observations underscore that the path to clinical translation requires not only biological rationale but also trial designs that incorporate predictive biomarkers, pharmacokinetic/pharmacodynamic assessments, and adaptive strategies to identify responsive patient subsets. CAF: cancer-associated fibroblast; CQ: chloroquine; PDTO: patient-derived tumor organoid; TME: tumor microenvironment.