CoChat
Sign Up
Cited by 201Gene expression and cancer classificationComputational Drug Discovery MethodsBreast Cancer Treatment Studies

The DOI matches its Crossref record; 57 of the 58 references it cites were found in a registry or an index, and 1 are not checked yet.

Predicting responses to platin chemotherapy agents with biochemically-inspired machine learning

Open access
Mucaki, Eliseos J., Zhao, Jonathan Z. L., Lizotte, Daniel J., Rogan, Peter K.·2019·Signal Transduction and Targeted Therapy
DOI 10.1038/s41392-018-0034-5PMC PMC6329797

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

The selection of effective genes that accurately predict chemotherapy responses might improve cancer outcomes. We compare optimized gene signatures for cisplatin, carboplatin, and oxaliplatin responses in the same cell lines and validate each signature using data from patients with cancer. Supervised support vector machine learning is used to derive gene sets whose expression is related to the cell line GI 50 values by backwards feature selection with cross-validation. Specific genes and functional pathways distinguishing sensitive from resistant cell lines are identified by contrasting signatures obtained at extreme and median GI 50 thresholds. Ensembles of gene signatures at different thresholds are combined to reduce the dependence on specific GI 50 values for predicting drug responses. The most accurate gene signatures for each platin are: cisplatin: BARD1 , BCL2 , BCL2L1 , CDKN2C , FAAP24 , FEN1 , MAP3K1 , MAPK13 , MAPK3 , NFKB1 , NFKB2 , SLC22A5 , SLC31A2 , TLR4 , and TWIST1 ; carboplatin: AKT1 , EIF3K , ERCC1 , GNGT1 , GSR , MTHFR , NEDD4L , NLRP1 , NRAS , RAF1 , SGK1 , TIGD1 , TP53 , VEGFB , and VEGFC; and oxaliplatin: BRAF , FCGR2A , IGF1 , MSH2 , NAGK , NFE2L2 , NQO1 , PANK3 , SLC47A1 , SLCO1B1 , and UGT1A1 . Data from The Cancer Genome Atlas (TCGA) patients with bladder, ovarian, and colorectal cancer were used to test the cisplatin, carboplatin, and oxaliplatin signatures, resulting in 71.0%, 60.2%, and 54.5% accuracies in predicting disease recurrence and 59%, 61%, and 72% accuracies in predicting remission, respectively. One cisplatin signature predicted 100% of recurrence in non-smoking patients with bladder cancer (57% disease-free; N = 19), and 79% recurrence in smokers (62% disease-free; N = 35). This approach should be adaptable to other studies of chemotherapy responses, regardless of the drug or cancer types.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41392-018-0034-5

Cites

57 of 58 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Schematic of platinum drug sensitivity and resistance genes that showed MFA correlation with the GI50 values for cisplatin. The gene products corresponding to those used to derive the SVM are indicated within boxes in the context of their cellular mechanisms of action and regulation of drug response. GE and CN correlations with inhibitory drug concentrations are based on the MFA of breast (GI50) and bladder (IC50) cancer cell line data. Gene products within the best-performing gene signature are embedded within color-coded ovals; whereas the other correlated gene products are embedded within rectangles
Fig. 1 Schematic of platinum drug sensitivity and resistance genes that showed MFA correlation with the GI50 values for cisplatin. The gene products corresponding to those used to derive the SVM are indicated within boxes in the context of their cellular mechanisms of action and regulation of drug response. GE and CN correlations with inhibitory drug concentrations are based on the MFA of breast (GI50) and bladder (IC50) cancer cell line data. Gene products within the best-performing gene signature are embedded within color-coded ovals; whereas the other correlated gene products are embedded within rectanglesCC BY · PubMed Central
Fig. 2 Schematic of platinum drug sensitivity and resistance genes that showed MFA correlation with the GI50 values for carboplatin. Refer to the legend of Fig. 1 for details
Fig. 2 Schematic of platinum drug sensitivity and resistance genes that showed MFA correlation with the GI50 values for carboplatin. Refer to the legend of Fig. 1 for detailsCC BY · PubMed Central
Fig. 3 Schematic of platinum drug sensitivity and resistance genes that showed MFA correlation with the GI50 values for oxaliplatin. Refer to the legend of Fig. 1 for details
Fig. 3 Schematic of platinum drug sensitivity and resistance genes that showed MFA correlation with the GI50 values for oxaliplatin. Refer to the legend of Fig. 1 for detailsCC BY · PubMed Central
Fig. 4 GI50 values for cell lines treated with the three platin drugs were plotted in order of ascending oxaliplatin GI50. For most cell lines, a trend between the GI50 values for cisplatin and carboplatin was observed, reflecting the correlation between the two drugs detected using MFA. Despite this correlation, carboplatin shows a much smaller variance (0.22) than cisplatin (0.37; the oxaliplatin variance is 0.34)
Fig. 4 GI50 values for cell lines treated with the three platin drugs were plotted in order of ascending oxaliplatin GI50. For most cell lines, a trend between the GI50 values for cisplatin and carboplatin was observed, reflecting the correlation between the two drugs detected using MFA. Despite this correlation, carboplatin shows a much smaller variance (0.22) than cisplatin (0.37; the oxaliplatin variance is 0.34)CC BY · PubMed Central
Fig. 5 Variation in the composition of the gene signatures obtained using misclassification-based SVMs at different GI50 thresholds for a cisplatin, b carboplatin, and c oxaliplatin. GI50 intervals are indicated on the left, with the number of cell lines with GI50 values within the indicated intervals shown in brackets. Each box represents the density of genes appearing in optimized Gaussian SVM gene signatures in those functional categories, with darker gray indicating frequent genes in the indicated GI50 threshold intervals and lighter gray indicating less commonly selected genes. The number of thresholded gene signatures used to derive the density plot within each interval is equal to (or greater than, in the case of multiple equally performing gene signatures) the number of cell lines within that GI50 interval
Fig. 5 Variation in the composition of the gene signatures obtained using misclassification-based SVMs at different GI50 thresholds for a cisplatin, b carboplatin, and c oxaliplatin. GI50 intervals are indicated on the left, with the number of cell lines with GI50 values within the indicated intervals shown in brackets. Each box represents the density of genes appearing in optimized Gaussian SVM gene signatures in those functional categories, with darker gray indicating frequent genes in the indicated GI50 threshold intervals and lighter gray indicating less commonly selected genes. The number of thresholded gene signatures used to derive the density plot within each interval is equal to (or greater than, in the case of multiple equally performing gene signatures) the number of cell lines within that GI50 intervalCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 57 references

Research works printed but not linked: 1

  1. Airley, R. Cancer Chemotherapy (Wiley-Blackwell, Hoboken, New Jersey 2009).

Cited by

Sorted by citations

Items 201–201 of 201

Related

See all

Metadata

Authors
Mucaki, E. J., Zhao, J. Z. L., Lizotte, D. J., & Rogan, P. K.
Affiliations
Western University (CA)
Corresponding author
Rogan, Peter K.
ORCID
Venue
Signal Transduction and Targeted Therapy
Publisher
Springer Science and Business Media LLC
Published
2019-01-11
Volume
4
Issue
1
Pages
1-1
ISSN
2059-3635 (ISSN-L)
DOI
10.1038/s41392-018-0034-5
PMC
PMC6329797
OpenAlex
W2950457484
Cited by
200
References
57
Citations by year
2026: 8 · 2025: 15 · 2024: 21 · 2023: 44 · 2022: 32 · 2021: 38 · 2020: 31 · 2019: 9
FWCI
9.10
Type
journal article
Language
en
Open access
Licence
Locations
  • Signal Transduction and Targeted Therapy (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Scholarship@Western (Western University) (acceptedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • Scholarship@Western (Western University) (submittedVersion)
Funders
  • Compute Canada
  • Canada Research Chairs
  • Natural Sciences and Engineering Research Council of Canada (RGPIN-2015, RGPIN-2015-06290)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Breast Cancer Treatment Studies
Keywords
cisplatin, carboplatin, oxaliplatin, drug response prediction, gene expression signatures, drug resistance, support vector machine, feature selection, bladder cancer, ovarian cancer, colorectal cancer, disease recurrence prediction, remission prediction

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review