An RNA damage response network mediates the lethality of 5-FU in colorectal cancer.

5-fluorouracil (5-FU), a major anti-cancer therapeutic, is believed to function primarily by inhibiting thymidylate synthase, depleting deoxythymidine triphosphate (dTTP), and causing DNA damage. Here, we show that clinical combinations of 5-FU with oxaliplatin or irinotecan show no synergy in human colorectal cancer (CRC) trials and sub-additive killing in CRC cell lines. Using selective 5-FU metabolites, phospho- and ubiquitin proteomics, and primary human CRC organoids, we demonstrate that 5-FU-mediated CRC cell killing primarily involves an RNA damage response during ribosome biogenesis, causing lysosomal degradation of damaged rRNAs and proteasomal degradation of ubiquitinated ribosomal proteins. Tumor types clinically responsive to 5-FU treatment show upregulated rRNA biogenesis while 5-FU clinically non-responsive tumor types do not, instead showing greater sensitivity to 5-FU's DNA damage effects. Finally, we show that treatments upregulating ribosome biogenesis, including KDM2A inhibition, promote RNA-dependent cell killing by 5-FU, demonstrating the potential for combinatorial targeting of this ribosomal RNA damage response for improved cancer therapy.

Additivity predicts the efficacy of most approved combination therapies for advanced cancer.

Most advanced cancers are treated with drug combinations. Rational design aims to identify synergistic combinations, but existing synergy metrics apply to preclinical, not clinical data. Here we propose a model of drug additivity for progression-free survival (PFS) to assess whether clinical efficacies of approved drug combinations are additive or synergistic. This model includes patient-to-patient variability in best single-drug response plus the weaker drug per patient. Among US Food and Drug Administration approvals of drug combinations for advanced cancers (1995-2020), 95% exhibited additive or less than additive effects on PFS times. Among positive or negative phase 3 trials published between 2014-2018, every combination that improved PFS was expected to succeed by additivity (100% sensitivity) and most failures were expected to fail (78% specificity). This study shows synergy is neither a necessary nor common property of clinically effective drug combinations. The predictable efficacy of approved combinations suggests that additivity can be a design principle for combination therapies.

An RNA Damage Response Network Mediates the Lethality of 5-FU in Clinically Relevant Tumor Types.

5-fluorouracil (5-FU) is a successful and broadly used anti-cancer therapeutic. A major mechanism of action of 5-FU is thought to be through thymidylate synthase (TYMS) inhibition resulting in dTTP depletion and activation of the DNA damage response. This suggests that 5-FU should synergize with other DNA damaging agents. However, we found that combinations of 5-FU and oxaliplatin or irinotecan failed to display any evidence of synergy in clinical trials, and resulted in sub-additive killing in a panel of colorectal cancer (CRC) cell lines. In seeking to understand this antagonism, we unexpectedly found that an RNA damage response during ribosome biogenesis dominates the drug's efficacy in tumor types for which 5-FU shows clinical benefit. 5-FU has an inherent bias for RNA incorporation, and blocking this greatly reduced drug-induced lethality, indicating that accumulation of damaged RNA is more deleterious than the lack of new RNA synthesis. Using 5-FU metabolites that specifically incorporate into either RNA or DNA revealed that CRC cell lines and patient-derived colorectal cancer organoids are inherently more sensitive to RNA damage. This difference held true in cell lines from other tissues in which 5-FU has shown clinical utility, whereas cell lines from tumor tissues that lack clinical 5-FU responsiveness typically showed greater sensitivity to the drug's DNA damage effects. Analysis of changes in the phosphoproteome and ubiquitinome shows RNA damage triggers the selective ubiquitination of multiple ribosomal proteins leading to autophagy-dependent rRNA catabolism and proteasome-dependent degradation of ubiquitinated ribosome proteins. Further, RNA damage response to 5-FU is selectively enhanced by compounds that promote ribosome biogenesis, such as KDM2A inhibitors. These results demonstrate the presence of a strong RNA damage response linked to apoptotic cell death, with clear utility of combinatorially targeting this response in cancer therapy.

Predictable Clinical Benefits without Evidence of Synergy in Trials of Combination Therapies with Immune-Checkpoint Inhibitors.

PURPOSE: Combinations of immune-checkpoint inhibitors (ICI) with other cancer therapies have been approved for advanced cancers in multiple indications, and numerous trials are under way to test new combinations. However, the mechanisms that account for the superiority of approved ICI combinations relative to their constituent monotherapies remain unknown. EXPERIMENTAL DESIGN: We analyzed 13 phase III clinical trials testing combinations of ICIs with each other or other drugs in patients with advanced melanoma and lung, breast, gastric, kidney, and head and neck cancers. The clinical activity of the individual constituent therapies, measured in the same or a closely matched trial cohort, was used to compute progression-free survival (PFS) curves expected under a model of independent drug action. To identify additive or synergistic efficacy, PFS expected under this null model was compared with observed PFS by Cox regression. RESULTS: PFS elicited by approved combination therapies with ICIs could be accurately predicted from monotherapy data using the independent drug action model (Pearson r = 0.98, P < 5 × 10-9, N = 4,173 patients, 8 types of cancer). We found no evidence of drug additivity or synergy except in one trial in which such interactions might have extended median PFS by 9 days. CONCLUSIONS: Combining ICIs with other cancer therapies affords predictable and clinically meaningful benefit by providing patients with multiple chances of response to a single agent. Conversely, there exists no evidence in phase III trials that other therapies interact with and enhance the activity of ICIs. These findings can inform the design and testing of new ICI combination therapies while emphasizing the importance of developing better predictors (biomarkers) of ICI response.

Aberrant Transcript Usage Is Associated with Homologous Recombination Deficiency and Predicts Therapeutic Response.

BRCA1/2 mutations account for only a small fraction of homologous recombination (HR) deficiency (HRD) cases. Recently developed genomic HRD (gHRD) tests suffer confounding factors that cause low precision in predicting samples that will respond to PARP inhibitors and DNA damaging agents. Here we present molecular and clinical evidence of transcriptional HRD (tHRD) that is based on aberrant transcript usage (aTU) of minor isoforms. Specifically, increased TU of nonfunctional isoforms of DNA repair genes was prevalent in breast and ovarian cancer with gHRD. Functional assays validated the association of aTU with impaired HR activity. Machine learning-based tHRD detection by the transcript usage (TU) pattern of key genes was superior to directly screening for gHRD or BRCA1/2 mutations in accurately predicting responses of cell lines and patients with cancer to PARP inhibitors and genotoxic drugs. This approach demonstrated the capability of tHRD status to reflect functional HR status, including in a cohort of olaparib-treated ovarian cancer with acquired platinum resistance. Diagnostic tests based on tHRD are expected to broaden the clinical utility of PARP inhibitors. SIGNIFICANCE: A novel but widespread transcriptional mechanism by which homologous recombination deficiency arises independently of BRCA1/2 mutations can be utilized as a companion diagnostic for PARP inhibitors.

Computational inference of cancer-specific vulnerabilities in clinical samples.

BACKGROUND: Systematic in vitro loss-of-function screens provide valuable resources that can facilitate the discovery of drugs targeting cancer vulnerabilities. RESULTS: We develop a deep learning-based method to predict tumor-specific vulnerabilities in patient samples by leveraging a wealth of in vitro screening data. Acquired dependencies of tumors are inferred in cases in which one allele is disrupted by inactivating mutations or in association with oncogenic mutations. Nucleocytoplasmic transport by Ran GTPase is identified as a common vulnerability in Her2-positive breast cancers. Vulnerability to loss of Ku70/80 is predicted for tumors that are defective in homologous recombination and rely on nonhomologous end joining for DNA repair. Our experimental validation for Ran, Ku70/80, and a proteasome subunit using patient-derived cells shows that they can be targeted specifically in particular tumors that are predicted to be dependent on them. CONCLUSION: This approach can be applied to facilitate the development of precision therapeutic targets for different tumors.