henhaobio
Sign in

Computational and Systems OncologyCancer BiologyPeer-reviewed journal article

Experiment-anchored abundance-coverage frontiers define conditional RHSVV exposure requirements in TP53-missense tumors

Takehiro IshikawaiD

PDFDOIPreprintCodeData
In-depth full-text digest available · Jump to digest ↓

Abstract

Mutant p53 accumulation is established, but whether abundance can complement conditional RHSVV/PAb240-like exposure depends on the trade-off between target coverage and normal-tissue separation. We analyzed 809 tumors and 523 organ-matched adjacent-normal specimens from eight CPTAC cohorts. The 67.5% high-p53 fraction among quantified DNA-binding-domain missense tumors and the 2.59-fold paired tumor/normal median were treated only as dataset-specific secondary calibration. The design pool comprised 316 annotation-compatible missense tumors; 268 had quantified p53. Eligibility retained 179 tumors at cohort-normal P95, 159 at R[≥]2, 90 at R[≥]3, 54 at R[≥]4, 21 at R[≥]6, and 8 at R[≥]8, where R is abundance relative to the cohort-normal median. Normal stress used a prespecified 2.1-fold total-p53 abundance multiplier from a vinculin-normalized, multi-time-point ionizing-radiation study in early-passage wild-type Mdm2+/+ mouse embryonic fibroblasts; this source met all primary-anchor eligibility criteria. Under this anchor and a sample-level exposed-state mixture weight p=0.90, the full-data lowest passing target/background exposure grid point was q=2.00 pooled and q=2.50 worst cohort at P95, versus q=1.25 for both at R[≥]3; R[≥]3 covered 33.6% and is presented as a pragmatic exploratory intermediate, not an optimum. In 500 whole-pipeline resamples, all seven fixed cohorts were evaluable at R[≥]3 in 416 (83.2%); this is an evaluability fraction, not design or treatment success. Separate 0.05-margin-buffered fixed R[≥]3 bootstrap candidates used q=1.50 pooled and q=1.75 worst cohort and met the internal constraints in 678/1,000 pooled resamples and 628/863 evaluable worst-cohort resamples (628/1,000 overall; 137 were not cohort-evaluable). These post-selection internal sampling audits are neither therapeutic probabilities nor validation. The principal result is an experiment-anchored coverage-q frontier specifying conditional experimental requirements; native RHSVV exposure, tumor killing, and safety remain unmeasured.

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (19 Aug 2026) · CC BY 4.0

Experiment-anchored sweep of p53 abundance thresholds in CPTAC TP53-missense tumors maps how much RHSVV exposure contrast is needed as target coverage falls.

Mutant p53 accumulation is known, but whether abundance can complement conformation-sensitive RHSVV exposure depends on separation from normal wild-type-p53 cells under stress. CPTAC lacks native RHSVV accessibility, so this work frames conditional exposure requirements instead of measuring them.

Key results

  1. ShownThe CPTAC release reproduces the known missense p53 accumulation signal. The authors treat this as dataset-specific secondary calibration, not a new mechanism.

    187/277 DBD-missense tumors (67.5%) above cohort-normal P95 vs 23/163 with no protein-altering mutation detected (14.1%) and 25/371 leave-one-out normals (6.7%); cohort-adjusted +0.831 log2, HC3 z=9.77, P=1.49×10^-22; paired median 2.59-fold (159 tumors) vs 1.09-fold (83). · Secondary dataset calibration reproduced the…

  2. ShownStricter abundance eligibility lowers coverage of the quantified molecular pool and erodes cohort representation. Landmarks R≥6 and R≥8 no longer retain the seven fixed reference cohorts.

    Of 316 compatible missense tumors, 268 had quantified p53. Retained tumors: 179 at P95 (66.8%), 159 at R≥2 (59.3%), 90 at R≥3 (33.6%), 54 at R≥4 (20.1%), 21 at R≥6 (7.8%), 8 at R≥8 (3.0%). R≥6 retained three cohorts, R≥8 one. · Prespecified abundance enrichment exposed a…

  3. SuggestedUnder the 2.1-fold normal stress anchor, the exposure contrast q needed to pass the response constraints falls as the abundance threshold rises, giving a coverage-q frontier. Abundance alone first passes at R≥4, covering 20.1% of the quantified pool.

    At p=0.90, lowest passing q (pooled/worst-cohort): P95 2.00/2.50; R≥2 1.75/2.00; R≥3 1.25/1.25; R≥4 1.00/1.00. q=1 failed at P95, R≥2 and R≥3; R≥4 covered 54 of 268 (20.1%). · Experiment-anchored abundance enrichment defined…; Conclusion

  4. SuggestedWhole-pipeline resampling shows the re-estimated q is lower and tighter at R≥3 than at P95. Seven-cohort evaluability at R≥3 is a representation fraction, not design success.

    500 resamples at p=0.90, median q (2.5th-97.5th pct): P95 2.25 (1.75-3.00) pooled and 2.75 (2.00-3.75) worst cohort; R≥3 1.50 (1.00-1.75) and 1.50 (1.25-2.00). All seven cohorts evaluable in 416/500 R≥3 resamples (83.2%). · Sampling audits separated q sensitivity from…

  5. SuggestedWith full-data R≥3 candidates held fixed, response constraints were met in a majority of resamples, but cohort attrition lowers the overall worst-cohort pass rate. These are internal post-selection audits, not validation.

    Pooled (q=1.50): 678/1,000 (67.8%; Wilson 64.8%-70.6%). Worst-cohort (q=1.75): 628/863 evaluable (72.8%; 69.7%-75.6%), equal to 628/1,000 overall with 137 cohort-attrition resamples counted as non-passes. · Sampling audits separated q sensitivity from…; Sampling analyses and reproducibility checks

Methods, data and reproducibility

CPTAC p53 is normalized to each cohort's adjacent-normal median to build a ladder of abundance eligibility thresholds. An exact monotone-cutpoint model then sweeps exposure mixture weight p and contrast q under a 2.1-fold normal stress multiplier, followed by two resampling audits.

  • Eight CPTAC cohorts (proteomics, mutation data) from LinkedOmicsKB Release 1, data freeze v1.2: tumors and organ-matched adjacent-normal specimens · human · n = 809 tumors; 523 adjacent-normal specimens
  • Quantified p53 subset, including patient-matched tumor/adjacent-normal pairs · human · n = 592 tumors; 371 adjacent normals; 327 matched pairs
  • Design pool of annotation-compatible TP53-missense tumors (RHSVV region sequence retained) · human · n = 316 tumors; 268 with quantified p53
  • Normal stress anchor: Pant-Lozano early-passage WT-MEF ionizing-radiation series (2.1-fold total p53 peak at 6 h after 6-Gy), used as the sole stress multiplier · mouse
  • Executable notebook, anchor documentation, derived frontier and sampling-audit outputs · 10.5281/zenodo.21964384
Methods & tools
CPTAC proteomics via LinkedOmicsKB Release 1, NCI Proteomic Data Commons (upstream resource), Exact empirical monotone-cutpoint analysis, Within-cohort bootstrap resampling (500 and 1,000 resamples), Wilson 95% confidence intervals, Mann-Whitney test, Cohort-adjusted linear model with HC3 standard errors, Cochran-Mantel-Haenszel test, Wilcoxon signed-rank test, Benjamini-Hochberg procedure
Reproducibility gaps
Software versions and compute are not reported. Full p-grid statuses with q and T values are in Supplementary Table S1, which is not part of the provided text.
Limitations (6)
  • The proteomic data capture total p53 only, with no readout of native RHSVV accessibility. Exposure parameters p and q are swept assumptions, and exposure is assumed independent of abundance because no joint distribution exists. · stated by the authors
  • The 2.1-fold stress anchor is a mouse-MEF value from one 6-Gy ionizing-radiation experiment, chosen for source fitness, not as universal or maximally conservative. Human normal cell types and clinical insults need direct measurement. · stated by the authors
  • Both sampling audits are internal post-selection checks on the same dataset, not held-out or external validation. All landmarks were analyzed on the same data. · stated by the authors
  • Missing p53 measurements affect 48 of 316 eligible missense tumors and 152 of 523 adjacent normals, so complete-case coverage may be biased by detection despite the reported bounds. Adjacent normal is a bulk organ-matched reference with possible field effects. · stated by the authors
  • The cohort minimum of three eligible tumors is small, leaving wide uncertainty in cohort-specific tails. The 5% and 10% worst-cohort normal limits are highly discrete in small cohorts. · stated by the authors
  • Only a single stress multiplier (2.1) was evaluated, so sensitivity of the frontier to alternative normal-stress magnitudes is not shown. Both sampling audits are also confined to p=0.90. · noted in this summary

Read this if you are weighing abundance-based indication thresholds for a conformation-specific p53 binder and want the coverage versus required exposure-contrast trade-off quantified.

Citations
0
FWCI
0.00
Field percentile
top 27%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISSadagopan '25Haratani '25Malhotra '23Loke '26Tracewell '26Bucho '22
44% semantic matchBiochemistry
Ananthan Sadagopan, Maximilian Carson, Eriks J. Zamurs+5 more·Nature Chemical Biology·Nov 3, 2025· 24 citations