Skip to main content

The disease burden of Ras

, by Ian A Prior

Ian Prior, Ph.D.

Ian Prior, Ph.D.

Ian Prior received his Ph. D. from the University of Liverpool where, after a postdoctoral interlude with John Hancock at the University of Queensland, he is now back as Head of the Department of Molecular Physiology and Cell Signalling, and North West Cancer Research Chair of Molecular Oncology.

I have always been struck by the lack of consensus around the mutation rates of Ras in cancer, with estimates ranging from ~10% to >30%. It seemed like something that we should know by now after such a long period of investigation and the large-scale cancer genetics databases that have been developed. There are similar discrepancies in the estimates for the mutation rates in common Ras cancers with pancreatic cancer ranging from 70% to 98% and lung and colorectal cancers ranging 25% to 50%. In a recently published article in Cancer Research [1], we explain the basis for these differences and suggest a way of more accurately estimating Ras disease burden.
    
cBioPortal, COSMIC, TCGA and ICGC databases and data portals are all excellent resources for investigating the mutation rates and other genetic changes in genes of interest across a wide spectrum of cancer types. Each database has particular strengths in terms of scale, consistency of curation, the range of genetic changes that have been mapped and the proportion of samples that have undergone genome-wide analysis. They each give a different value of (H+K+N) Ras mutation rates ranging from ~12% in TCGA to ~25% in COSMIC. A major reason for these differences is due to the different compositions of the databases; those that contain more samples from common KRAS mutant cancers have higher estimates. Importantly, the compositions of the databases do not reflect that relative incidence of each cancer type within the population, meaning that their top-level statistics do not directly estimate disease burden. Individual cancer types also show a range of estimates for a given Ras gene; for example, KRAS mutation rates in pancreatic cancer varied from 58% to 80% across the databases. None of these values are close to the >90% mutation frequency that we all typically cite. The private Foundation Medicine (FM) database suggests a reason for this discrepancy; when pancreatic cancer samples are curated to exclude those with significant stromal content, the KRAS mutation rate is ~88% [2].

In order to try to develop a consensus for Ras disease burden, we converted the Ras mutation frequencies of the largest databases (COSMIC and publicly available FM data) into patient numbers for each cancer type based on current American Cancer Society disease incidence statistics. We found that ~19% of all US cancer patients habour a Ras mutation, equivalent to ~250,000 new cases per year in the United States. There are caveats with extrapolating this US-based estimate into a global context due the different geographical frequencies of distinct cancer types; however, 19% of global cancer cases is equivalent to ~3.5 million new Ras mutant cancer patients per year.

The top 20 cancers in the United States in terms of the number of Ras mutant patients can be seen in Table 1. A feature of this type of analysis is that it highlights the significant pools of patients that may benefit from some of the current Ras-relevant clinical trials. For example, ongoing HRAS directed clinical trials of farnesyltransferase inhibitor therapies have focussed on cancer types where HRAS mutation frequencies are relatively high such as head and neck blood and thyroid cancers. In fact, the ~1,600 HRAS mutant US breast cancer patients translate into ~12,000 cases globally who might also benefit from these treatments in a disease not typically thought to be so directly Ras relevant.

Whilst this analysis focussed on the Ras cancer disease burden, Ras contributions to RASopathies are also important. Genetic databases for these diseases are now also being developed, with the NSEuroNet database coordinated by Martin Zenker representing a leading example. Applying a similar approach of converting mutation frequencies into patient numbers estimates that ~5% of RASopathy patients will harbour a mutation in one of the Ras isoforms, equivalent to up to 400,000 people worldwide. Interesting isoform and mutation-specific patterns are also becoming evident within the NSEuroNet database that underscore the complexity of Ras variant biology across all Ras mutant diseases.

Hopefully these numbers provide some food for thought. At the very least I now know what percentage I’m going to quote when I next apply for funding.

The Top 20 Most Frequently Observed Ras Mutant Cancers in the United States
Cancer Types Tissue Total Cases/Year % Mutated Estimated New Patients/Year (USA)
      H K N HRAS KRAS NRAS RAS
COAD colon 97,220 0.5 50.0 4.2 486 48,610 4,083 53,179
PAAD pancreas 55,440 0.0 88.0 0.4 0 48,787 238 49,026
LUAD lung 93,618 0.2 32.0 1.0 183 29,956 964 31,103
READ rectum 51,610 0.2 50.0 4.1 103 25,805 2,116 28,024
SKCM skin 91,270 1.2 1.6 16.9 1,136 1,504 15,428 18,067
UCEC uterus 60,701 0.2 16.8 2.8 147 10,224 1,671 12,042
PCM blood 30,770 0.8 17.7 19.3 252 5,444 5,944 11,639
BLCA bladder 81,190 7.0 5.0 1.4 5,670 4,083 1,096 10,849
PRAD prostate 164,690 1.5 3.4 0.3 2,464 5,618 488 8,570
BRCA breast 268,670 0.6 1.3 0.4 1,613 3,578 1,001 6,193
HNSC head & neck 64,690 5.1 2.0 1.6 3,320 1,272 1,054 5,646
THCA thyroid 42,323 2.2 1.7 5.9 927 704 2,494 4,125
AML blood 19,520 0.0 5.3 14.3 5 1,025 2,793 3,822
LUSC lung 70,209 0.6 4.0 0.6 441 2,794 402 3,637
SIAD small intestine 10,470 0.0 26.4 1.0 0 2,763 100 2,863
OV ovary 22,240 0.2 8.6 1.6 47 1,914 361 2,323
GBC gall bladder 12,190 0.8 16.1 1.6 101 1,964 195 2,259
THCAF thyroid 5,643 7.0 4.6 18.9 393 260 1,065 1,718
DLBC blood 74,680 0.0 1.3 0.6 0 967 458 1,425
CML blood 8,430 0.3 6.4 9.7 26 538 817 1,382
AML, acute myeloid leukemia; BLCA, bladder urothelial carcinoma; BRCA, breast invasive carcinoma; CML, chronic myeloid leukaemia; COAD, colon adenocarcinoma; DLBC, non-Hodgkin lymphoma; GBC, gallbladder carcinoma; HNSC, head and neck squamous cell carcinoma; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; OV, ovarian serous cystadenocarcinoma; PAAD, pancreatic adenocarcinoma; PCM, plasma cell myeloma; READ, rectal adenocarcinoma; SIAD, small intestine adenocarcinoma; SKCM, skin cutaneous melanoma; THCA, papillary thyroid carcinoma; THCAF, follicular thyroid carcinoma; UCEC, uterine corpus endometrial carcinoma.
Selected References
  1. Prior IA, Hood FE and Hartley JL, 2020. The frequency of Ras mutations in cancer. Cancer Research.

    [PubMed Abstract]
  2. Singhi AD, George B, Greenbowe JR, Chung J, Suh J, Maitra A, Klempner SJ, Hendifar A, Milind JM, Golan T, Brand RE, Zureikat AH, Roy S, Schrock AB, Miller VA, Ross JS, Ali SM, Bahary N, 2019. Real-Time Targeted Genome Profile Analysis of Pancreatic Ductal Adenocarcinomas Identifies Genetic Alterations That Might Be Targeted With Existing Drugs or Used as Biomarkers. Gastroenterology

    [PubMed Abstract]
< Older Post

Alternative cells-of-origin influence the heterogeneity of KRAS-mutant lung adenocarcinoma

Newer Post >

RAS Mutation Tropism

If you would like to reproduce some or all of this content, see Reuse of NCI Information for guidance about copyright and permissions. In the case of permitted digital reproduction, please credit the National Cancer Institute as the source and link to the original NCI product using the original product's title; e.g., “The disease burden of Ras was originally published by the National Cancer Institute.”

We welcome your comments on this post. All comments must follow our comment policy.

Archive