Testing patterns, patient and tumour characteristics and survival by NRAS and KIT genotype in melanoma
Khaylen Mistry, Polly Jeffrey, Nick J Levell, Oliver Kennedy, Kathryn Richardson, Paul Craig, Chloe Bright, Siobhan Taylor, John Ragan, Dimitrios Karponis, Joanna Pethick, Katrina Lavelle, Fiona McRonald, Steven Hardy, Sally Vernon, Robert Dawe, Charlotte Proby, Paul Lorigan
TL;DR
This study analyzes the frequency and impact of NRAS and KIT mutations in melanoma patients in England, highlighting geographic and demographic variations in testing and the need for targeted therapies.
Contribution
The study presents the largest national dataset on melanoma NRAS and KIT status, revealing regional and demographic testing disparities and emphasizing the need for NRAS-targeted treatments.
Findings
NRAS and KIT mutations were tested in 6.6% and 3.0% of melanomas, respectively, with significant regional and demographic variations in testing rates.
NRAS mutations were more common in older patients and less common in women and those with head/neck melanomas.
No significant difference in 5-year survival was found between NRAS/KIT wildtype and mutated tumors.
Abstract
NRAS and KIT mutations in melanoma bring implications for prognosis, follow-up, selection into trials and potential future treatment with targeted therapies. The frequency of NRAS/KIT mutations and their association with patient/tumour characteristics and survival is poorly documented. To report national data from England on (i) the frequency of NRAS and KIT mutations, (ii) the association of patient/tumour characteristics with NRAS and KIT mutations, and (iii) the survival of patients with NRAS and KIT mutations. This retrospective cohort study identified all new melanomas diagnosed in England from 2016 to 2021 and molecular NRAS/KIT testing using data from the National Disease Registration Service. Multivariate logistic regression determined the association between (i) NRAS and KIT testing and patient/tumour characteristics, and (ii) NRAS genotype and patient/tumour characteristics.…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
| Variable | WT ( | Mutated ( | Total ( | Logistic regression, ORa (95% CI) ( | |
|---|---|---|---|---|---|
| Univariate analysis | Multivariate analysis | ||||
| Gender | |||||
| Men | 2290 (59.0) | 1059 (60.4) | 3349 (59.4) | REF | REF |
| Women | 1590 (41.0) | 695 (39.6) | 2285 (40.6) | 0.95 (0.84–1.06) |
|
| Age (years), median (IQR) | 70 (48–92) | 72 (54–90) | 71 (50–92) |
|
|
| Siteb | |||||
| Skin | 3880 (95.2) | 1754 (96.9) | 5634 (95.7) | REF | REF |
| Mucosal | 142 (3.5) | 52 (2.9) | 194 (3.3) | 0.79 (0.56–1.09) | 0.81 (0.57–1.13) |
| Other | 54 (1.3) | 5 (0.3) | 59 (1.0) |
|
|
| Skin site | |||||
| Head/neck | 842 (21.7) | 226 (12.9) | 1068 (19.0) |
|
|
| Limbs | 1468 (37.8) | 893 (50.9) | 2361 (41.9) | REF | REF |
| Trunk | 1100 (28.4) | 445 (25.4) | 1545 (27.4) |
|
|
| Overlapping/unknown/external genitals | 470 (12.1) | 190 (10.8) | 660 (11.7) |
|
|
| Ethnicityc | |||||
| White | 3607 (93.0) | 1641 (93.6) | 5248 (93.1) | REF | REF |
| Minority ethnic grouping | 93 (2.4) | 29 (1.7) | 122 (2.2) | 0.69 (0.44–1.03) | 0.74 (0.47–1.13) |
| Black | 13 (0.3) | 4 (0.2) | 17 (0.3) | – |
|
| Asian | 19 (0.5) | 8 (0.5) | 27 (0.5) | – |
|
| Mixed | 7 (0.2) | 2 (0.1) | 9 (0.2) | – |
|
| Other | 54 (1.4) | 15 (0.9) | 69 (1.2) | ||
| Unknown | 180 (4.6) | 84 (4.8) | 264 (4.7) | 1.03 (0.78–1.33) | 1.07 (0.81–1.40) |
| Deprivation quintile | |||||
| 1 (most deprived) | 397 (10.2) | 157 (9.0) | 554 (9.8) | 0.82 (0.66–1.01) | 0.85 (0.68–1.06) |
| 2 | 583 (15.0) | 262 (14.9) | 845 (15.0) | 0.93 (0.77–1.11) | 0.93 (0.77–1.12) |
| 3 | 927 (23.9) | 391 (22.3) | 1318 (23.4) | 0.87 (0.74–1.02) | 0.85 (0.72–1.00) |
| 4 | 946 (24.4) | 447 (25.5) | 1393 (24.7) | 0.98 (0.84–1.14) | 0.96 (0.82–1.13) |
| 5 (least deprived) | 1027 (26.5) | 497 (28.3) | 1524 (27.1) | REF | REF |
| Stage | |||||
| I | 412 (10.6) | 162 (9.2) | 574 (10.2) |
| 0.81 (0.66–1.00) |
| IA | 98 (2.5) | 33 (1.9) | 131 (2.3) | – | – |
| IB | 309 (8.0) | 122 (7.0) | 431 (7.6) | – | – |
| IX (stage I but not specified if IA/B) | 5 (0.1) | 7 (0.4) | 12 (0.2) | – | – |
| II | 1325 (34.1) | 676 (38.5) | 2001 (35.5) | REF | REF |
| IIA | 271 (7.0) | 178 (10.1) | 449 (8.0) | – | – |
| IIB | 464 (12.0) | 269 (15.3) | 733 (13.0) | – | – |
| IIC | 584 (15.1) | 228 (13.0) | 812 (14.4) | – | – |
| IIX (stage II but not specified if IIA/B/C) | 6 (0.2) | <5 | 7 (0.1) | – | – |
| III | 986 (25.4) | 411 (23.4) | 1397 (24.8) |
|
|
| IIIA | 123 (3.2) | 28 (1.6) | 151 (2.7) | – | – |
| IIIB | 194 (5.0) | 94 (5.4) | 288 (5.1) | – | – |
| IIIC | 408 (10.5) | 186 (10.6) | 594 (10.5) | – | – |
| IIID | 39 (1.0) | 8 (0.5) | 47 (0.8) | – | – |
| IIIX (stage III but not specified if IIIA/B/C/D) | 222 (5.7) | 95 (5.4) | 317 (5.6) | – | – |
| IV | 394 (10.2) | 167 (9.5) | 561 (10.0) | 0.83 (0.68–1.02) | 0.87 (0.70–1.08) |
| IVX | 394 (10.2) | 167 (9.5) | 561 (10.0) | – | – |
| Unknown | 763 (19.7) | 338 (19.3) | 1101 (19.5) | 0.87 (0.74–1.02) | 0.90 (0.76–1.08) |
| Region | |||||
| London | 433 (11.2) | 195 (11.1) | 628 (11.1) | 0.98 (0.80–1.19) | 1.03 (0.84–1.27) |
| East of England | 771 (19.9) | 343 (19.6) | 1114 (19.8) | 0.97 (0.82–1.14) | 1.01 (0.85–1.20) |
| North East | 292 (7.5) | 121 (6.9) | 413 (7.3) | 0.90 (0.71–1.14) | 0.98 (0.77–1.25) |
| North West | 114 (2.9) | 58 (3.3) | 172 (3.1) | 1.10 (0.79–1.53) | 1.23 (0.87–1.72) |
| Yorkshire and the Humber | 328 (8.5) | 135 (7.7) | 463 (8.2) | 0.89 (0.71–1.12) | 0.99 (0.84–1.94) |
| East Midlands | 70 (1.8) | 37 (2.1) | 107 (1.9) | 1.15 (0.75–1.72) | 1.28 (0.84–1.94) |
| West Midlands | 244 (6.3) | 131 (7.5) | 375 (6.7) | 1.16 (0.92–1.48) |
|
| South East | 1063 (27.4) | 490 (27.9) | 1553 (27.6) | REF | REF |
| South West | 565 (14.6) | 244 (13.9) | 809 (14.4) | 0.94 (0.78–1.13) | 0.96 (0.79–1.16) |
| Stage, genotype, gender and age status | Total | 5-year NSa (95% CI) |
|---|---|---|
|
| 4050 | |
| All stages | 2772 | 62.3 (58.9–65.9) |
| All stages | 1278 | 58.5 (54.0–63.4) |
| Stage I | 286 | 70.0 (59.2–82.8) |
| Stage II | 112 | 27.9 (19.7–39.6) |
| Stage II | 1038 | 64.1 (58.7–70.0) |
| Stage II | 552 | 68.3 (62.1–75.1) |
| Stage IIA | 223 | 71.3 (61.2–82.9) |
| Stage IIA | 157 | 64.1 (52.3–78.7) |
| Stage IIB | 360 | 63.4 (53.3–75.5) |
| Stage IIB | 214 | 74.0 (64.2–85.3) |
| Stage IIC | 453 | 58.8 (50.8–68.0) |
| Stage IIC | 180 | 57.1 (45.8–71.2) |
| Stage III | 762 | 62.0 (55.6–69.2) |
| Stage III | 309 | 54.2 (45.6–64.5) |
| Stage IV | 302 | 33.2 (26.3–42.0) |
| Stage IV | 120 | 33.5 (24.2–46.3) |
| Stage unknown | 384 | 71.5 (62.9–81.3) |
| Stage unknown | 185 | 48.5 (34.3–68.5) |
|
| 1482 | |
| All stages | 1417 | 50.4 (44.4–57.3) |
| All stages | 65 | 52.1 (37.1–73.2) |
| Variables | Wildtype ( | Mutated ( | Total ( |
|---|---|---|---|
| Gender | |||
| Men | 1243 (59.7) | 53 (56) | 1296 (59.5) |
| Women | 840 (40.3) | 42 (44) | 882 (40.5) |
| Age (years), median (IQR) | 70 (49–91) | 75 (60–90) | 70 (49–91) |
| Sitea | |||
| Skin | 2083 (86.4) | 95 (64) | 2178 (85.1) |
| Mucosal | 276 (11.4) | 50 (34) | 326 (12.7) |
| Other | 53 (2.2) | < 5 | 56 (2.2) |
| Skin site | |||
| Head/neck | 381 (18.3) | 26 (27) | 407 (18.7) |
| Limbs | 919 (44.1) | 46 (48) | 965 (44.3) |
| Trunk | 535 (25.7) | 14 (15) | 549 (25.2) |
| Overlapping/unknown | 248 (11.9) | 9 (9) | 257 (11.8) |
| Ethnicityb | |||
| White | 1901 (91.3) | 84 (88) | 1985 (91.1) |
| Minority ethnic grouping | 68 (3.3) | 7 (7) | 75 (3.4) |
| Unknown | 114 (5.5) | < 5 | 118 (5.4) |
| Deprivation quintile | |||
| 1 (most deprived) | 193 (9.3) | 10 (11) | 203 (9.3) |
| 2 | 310 (14.9) | 15 (16) | 325 (14.9) |
| 3 | 517 (24.8) | 24 (25) | 541 (24.8) |
| 4 | 509 (24.4) | 27 (28) | 536 (24.6) |
| 5 (least deprived) | 554 (26.6) | 19 (20) | 573 (26.3) |
| Stage | |||
| I | 179 (8.6) | 5 (5) | 184 (8.4) |
| II | 702 (33.7) | 39 (41) | 741 (34.0) |
| III | 529 (25.4) | 22 (23) | 551 (25.3) |
| IV | 235 (11.3) | 11 (12) | 246 (11.3) |
| Unknown | 438 (21.0) | 18 (19) | 456 (20.9) |
| Region | |||
| London | 365 (17.5) | 20 (21) | 385 (17.7) |
| East of England | 514 (24.7) | 22 (23) | 536 (24.6) |
| North East | 29 (1.4) | < 5 | 29 (1.3) |
| North West | 57 (2.7) | 5 (5) | 62 (2.8) |
| Yorkshire and the Humber | 88 (4.2) | 6 (6) | 94 (4.3) |
| East Midlands | 37 (1.8) | < 5 | 37 (1.7) |
| West Midlands | 229 (11.0) | 8 (8) | 237 (10.9) |
| South East | 352 (16.9) | 19 (20) | 371 (17.0) |
| South West | 412 (19.8) | 15 (16) | 427 (19.6) |
| Management | |||
| Definitive surgery | 1806 (86.7) | 77 (81) | 1883 (86.5) |
| Systemic therapy | 830 (39.8) | 34 (36) | 864 (39.7) |
| Radiotherapy | 196 (9.4) | 10 (11) | 206 (9.5) |
| Systemic therapyc | |||
| Immunotherapy only | 815 (40.2) | 36 (39) | 851 (40.2) |
| Targeted therapy only | 168 (8.3) | < 5 | 170 (8.0) |
| Targeted and immunotherapy | 122 (6.0) | 5 (5) | 127 (6.0) |
| Neither | 920 (45.4) | 49 (53) | 969 (45.8) |
- —Cancer Research UK10.13039/501100000289
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsCutaneous Melanoma Detection and Management · Melanoma and MAPK Pathways · melanin and skin pigmentation
Mutations in the BRAF, NRAS and KIT genes are important oncogenic ‘drivers’ of melanoma. Since 2013, international guidance and expert consensus have recommended routinely determining BRAF status in stage III and IV melanoma, and some recommendations include stage II disease.^1–5^ BRAF mutations are present in approximately 30–50% of cutaneous melanomas, are more common in certain demographics such as younger patients and are prognostically significant and predictive of response to targeted therapy and immunotherapy in the adjuvant and palliative settings.^6–9^
By contrast, the frequency of NRAS and KIT mutations and their association with patient and tumour characteristics and survival are less well understood.^10,11^ Many previous epidemiology studies have been limited by small cohorts, with a lack of comprehensive data on a range of key covariates, and many studies pooled data for various cancers.^10^ NRAS and KIT mutations are important beyond their frequency because of their implications for prognosis, follow-up recommendations, selection into clinical trials and the unmet need for novel targeted therapies.
In the UK, there are no national guidelines for NRAS/KIT testing in specific melanoma subtypes. Current European guidelines recommend testing for NRAS and KIT mutations if the tumour is BRAF wildtype (WT) at the V600 locus.^4^ Many UK pathology laboratories do not routinely test for NRAS/KIT mutations, as these are not actionable mutations and most are selective about which melanoma types and stages they will request testing on. Sometimes NRAS testing is requested for more advanced stage melanoma and KIT testing for mucosal or acral lentiginous melanoma. The National Genomic Test Directory for Cancer in England was first published in 2018; its implementation drives access for all patients to next-generation sequencing panels that may provide greater information on the incidence of NRAS and KIT mutations.^12^
No national melanoma NRAS and KIT genetics data from England have been reported, to the best of our knowledge. The National Institute for Health and Care Excellence has recommended national cohort studies to evaluate melanoma biomarkers in treatment planning, response to treatment and prognosis.^1^ The aim of this study was to report national data for diagnoses of melanoma from England between 2016 and 2021 on (i) the frequency of molecular NRAS and KIT mutations, (ii) the association of patient demographics and tumour characteristics with NRAS and KIT mutations and (iii) the survival of patients with NRAS and KIT mutations.
Patients and methods
Study design, data sources and variables
This national cohort study used English cancer registry data from the National Disease Registration Service (NDRS).^13^ The registry maintains details of all cancers diagnosed each year across England (population 56 489 800 in the 2021 census).^14^ It is mandatory for all National Health Service (NHS) pathology laboratories, and recommended for all private pathology laboratories in England to provide all cancer pathology reports to the dataset. Only a small proportion are managed in the private sector in England and NDRS data are considered the gold-standard dataset for cancer data representation in England. Pathology report data are combined with the Patient Administration System and Cancer Outcomes and Services Dataset to form the National Cancer Registration Dataset (NCRD).^14^ The NDRS receives and records data on molecular genetic testing (including NRAS and KIT mutational analysis) directly from genomics or molecular pathology laboratories across England.
Patient records within the NCRD were linked to other datasets, including the National Radiotherapy Dataset, the Systemic Anti-Cancer Therapy (SACT) dataset, the NHS Hospital Episode Statistics datasets and death registrations from the Office for National Statistics.^15–17^ These linked data sources provided information on genetics and treatment pathways, including systemic therapy, radiotherapy and operation code data.
Data were extracted on 1 February 2024 for new melanomas diagnosed from 1 January 2016 to 31 December 2021, which was the latest available complete year. Melanoma was identified using International Classification of Diseases (ICD)-10 site codes and ICD-O-3 morphology and behaviour codes (Table S1; see Supporting Information).^18^ Disease-specific death was defined by the ICD-10 code C43 or C80 (melanoma skin cancer, malignant neoplasm without specification of site). Cause of death and vital status of patients was determined until 5 July 2023.
Patient variables extracted included age at diagnosis, gender, ethnicity, tumour site, stage at diagnosis, geography and deprivation. Table S2 (see Supporting Information) describes how variables were defined. Owing to low numbers for analysis, ethnicity was grouped into White and a minority ethnic grouping including Black, Asian, Chinese, Mixed and Other as per the Office for National Statistics categorization (see Table S2 for further information). REporting of studies Conducted using Observational Routinely-collected Data (RECORD) guidelines were adhered to (Appendix S1; see Supporting Information).
Outcomes and statistical analysis
Data were extracted using Oracle SQL developer 19.4.0.354.1759. Statistical analyses were completed using R 4.3.2 and Stata 18.0. Multivariate logistic regression examined the association between NRAS and KIT testing of melanoma and covariates to understand any potential selection bias.
Multivariate logistic regression was used to examine the association between NRAS and KIT genotypes with covariates. The variance inflation factor was checked to ensure minimal multicollinearity. Secondary logistic regression evaluated whether tumours that were NRAS tested > 90 days from the pathological diagnosis date were more likely to be NRAS mutated, which may have suggested that patients with recurrent disease were more likely to be positive.
Survival time was calculated for patients diagnosed with their first malignant melanoma in England between 2016 and 2020. Patients were censored at the first occurrence of death, loss to follow-up (loss to NHS through lack of updated information or emigration) or the study end date (5 July 2023). The Pohar Perme estimator was used to calculate 5-year age-standardized net survival (NS) estimates by comparing the survival of patients with melanoma with that expected based on the general population of the same profile of age (single year), gender, deprivation and geographic region.^19–22^ Hazard ratios (HR) for disease-specific mortality were estimated using univariate and multivariate cause-specific Cox regression models, with a second multivariate model including SACT limited to stage III/IV melanomas, with time to melanoma-specific death as the outcome. The proportional hazards assumption was examined for each covariate using log–log plots.
Sensitivity analysis to evaluate potential immortal time bias compared the HRs from the multivariate Cox models starting from the diagnosis date and NRAS/KIT testing date (cohort restricted to pathological diagnosis date – NRAS/KIT test date ≤ 90 days).
Results
NRAS testing patterns
Of new melanomas diagnosed, 6.6% (6045/91 415) had an NRAS test registered and 2.6% (158/6045) of those tested had either a failed, inconclusive or unknown test result. Of the successfully tested tumours, 95.7% (5634/5887) were cutaneous. The proportion of successfully tested tumours across all sites and cutaneous sites that were NRAS mutated was 30.8% (1811/5887) and 31.1% (1754/5634), respectively.
The percentage of stage III or IV cutaneous tumours that were NRAS tested was 22.4% (1958/8731), whereas the percentage of stage II and I tumours tested was 11.9% (2001/16 844) and 1.1% (574/52 353), respectively (Table S3; see Supporting Information). The East of England tested the highest proportion of cutaneous tumours, 11.2% (1114/9950), compared with the lowest in the North West, 1.4% (172/12 296).
Women [odds ratio (OR) 0.81, 95% confidence interval (CI) 0.76–0.86] were less likely to be NRAS tested. There was no association between age and NRAS testing. Mucosal melanomas (OR 2.08, 95% CI 1.73–2.49), melanomas at overlapping/unknown/external genital sites (OR 1.80, 95% CI 1.61–2.01), patients in the minority ethnic grouping (OR 1.31, 95% CI 1.06–1.62) and more deprived individuals (OR 1.12, 95% CI 1.01–1.25) were more likely to be NRAS tested.
Patient and tumour characteristics and treatment received by NRAS genotype
In those tested, the study found there was a lower odds of having an NRAS mutation associated with patients who were women (OR 0.81, 95% CI 0.71–0.91) (Table 1). The NRAS-mutated genotype was associated with older age (OR 1.01, 95% CI 1.01–1.02). For cutaneous melanoma, the head/neck had the lowest odds of being NRAS mutated compared with the reference site of the limbs (OR 0.37, 95% CI 0.31–0.44) (Table 1). There was no association between whether a tumour was NRAS tested after 90 days from pathological diagnosis and NRAS genotype (OR 1.02, 95% CI 0.90–1.16) (Table S4; see Supporting Information).
Among patients with an NRAS mutation, 47.4% (805/1697) received immunotherapy only, 0.7% (12/1697) received targeted therapy only, 0.5% (9/1697) received both immunotherapy and targeted therapy and 51.3% (871/1697) received neither (Table S5; see Supporting Information).
Survival by NRAS genotype
There was no significant difference in 5-year NS between people with all-stage NRAS WT and those with mutated tumours (WT NS 62.3%, 95% CI 58.9–65.9 vs. mutated NS 58.5%, 95% CI 54.0–63.4) (Table 2). Stage I NRAS-mutated genotype was associated with a lower 5-year NS (WT NS 70.0%, 95% CI 59.2–82.8 vs. mutated NS 27.9%, 95% CI 19.7–39.6).
Similarly, in the multivariate Cox model, the NRAS-mutated genotype was not associated with a significant difference in disease-specific mortality (HR 1.12, 95% CI 1.00–1.25) (Table S6; see Supporting Information). SACT was not associated with disease-specific mortality in melanomas that underwent NRAS testing (HR 1.05, 95% CI 0.88–1.26) (Table S7; see Supporting Information). HRs from multivariate Cox models measuring survival from the diagnosis date or the NRAS testing date were similar, confirming minimal immortal time bias (HR 1.10, 95% CI 0.94–1.29) (Table S8; see Supporting Information).
KIT testing patterns, characteristics and survival
Table 3 details the patient and tumour characteristics by KIT genotype. Of new melanomas diagnosed, 3.0% (2705/91 415) had a KIT test registered, of which 5.4% (145/2705) had either a failed, inconclusive or unknown test result. Of the successfully tested tumours, 85.1% (2178/2560) were cutaneous and 12.7% (326/2560) were mucosal (including external genital). The proportion of successfully tested tumours across all sites, cutaneous sites and mucosal sites (including external genital) that were KIT mutated was 5.8% (148/2560), 4.4% (95/2178) and 15.3% (50/326), respectively. The percentage of stage III or IV cutaneous tumours that were KIT tested was 9.2% (797/8640), whereas the percentage of stage II and I tumours tested was 4.4% (741/16 692) and 0.4% (184/52 289), respectively (Table S9; see Supporting Information).
London tested the highest proportion of cutaneous tumours (5.4%, 385/7175) compared with the lowest in the North West (0.5%, 62/12 253). Patients who were women (OR 0.83, 95% CI 0.76–0.92) were less likely to be KIT tested. Mucosal (OR 8.04, 95% CI 5.84–11.03), cutaneous melanomas at overlapping or unknown sites (OR 1.61, 95% CI 1.36–1.90) and patients in the minority ethnic grouping (OR 1.78, 95% CI 1.37–2.30) were more likely to be KIT tested. A similar proportion of women who were KIT tested were KIT mutant compared with men [4.8% (42/882) vs. 4.1% (53/1296)].
There was no significant difference in 5-year NS between people with KIT WT and those with mutated tumours (KIT WT 5-year NS 50.4%, 95% CI 44.7–57.3 vs. KIT-mutated 5-year NS 52.1%, 95% CI 37.1–73.2) (Table 2).
Discussion
This study presents the largest dataset on national melanoma NRAS and KIT genotype status published to date and the most complete reporting on testing patterns, characteristics and survival associated with NRAS and KIT genotypes, to the best of our knowledge. High-quality research has identified the need for detection of BRAF, NRAS and KIT mutations for metastatic or advanced acral and nail apparatus melanoma to provide the best therapeutic approach for patients.^23,24^
Marked geographical variation in NRAS and KIT testing across regions of England was observed, which may reflect the lack of national guidance for testing during the study period. Other explanations include regional differences in population characteristics, selective testing and variation in compliance in sending data between regional laboratories. Geographical testing patterns appeared similar for NRAS and KIT; therefore, variations in mutation testing methods (specific NRAS tests vs. panel testing with BRAF/NRAS/KIT) may have influenced results. Both NRAS and KIT testing were more common in patients who were men, had advanced melanoma, had mucosal melanoma, had melanoma in overlapping/unknown/external genital cutaneous sites and were in the minority ethnic grouping. Male patients and minority ethnic groups have been found to present with later stage melanoma, which may explain the higher testing.^10^ KIT mutations are associated with mucosal and acral melanoma and the greater testing in minority ethnic groups may be explained by the higher proportion of acral melanoma; however, no data were available for melanoma subtype.^10,25,26^
NRAS mutations were identified in 31.1% (1754/5634) of cutaneous tumours tested, almost double that of a recent global meta-analysis that reported 16.4% (641/3904). This discrepancy may reflect selection bias towards higher-risk tumours being tested.^10^ Another recent NDRS publication reported 35.2% (4401/12 490) of cutaneous tumours tested were BRAF mutated, which is lower than that routinely reported in the literature.^9^ These findings that NRAS-mutated (31.1%) and BRAF-mutated (35.2%) melanomas have a broadly similar incidence in melanomas undergoing testing in clinical practice are essential for clinical trial development as well as updating medical literature.^9^ As both mutations activate mitogen-activated protein kinase signalling, their mutual exclusivity was expected and confirmed; 97.3% (1745/1794) of NRAS-mutated tumours were BRAF WT.
NRAS mutations for all-stage melanoma were not significantly associated with survival. This is the case despite the absence of NRAS-directed therapies, in contrast with BRAF. The lower survival observed in stage I NRAS-mutated melanoma should be interpreted with caution given possible selective testing of ulcerated stage IB disease, the small sample size of stage I NRAS-tested melanoma and potential NRAS testing on recurrence. This dataset only recorded tumour stage at diagnosis and there was no validated marker for recurrence; hence, molecular testing may have occurred on disease recurrence. However, no association was found between a longer time lag between diagnosis date and genetic test date and NRAS genotype. Previous literature has associated NRAS-mutated melanoma with a higher tumour mutation burden and greater risk of brain metastases.^27,28^ NRAS mutations may be associated with a modest response to mitogen-activated protein kinase inhibitors, but the lack of specific NRAS-targeted therapies is an area of unmet need, particularly for advanced melanoma that has not responded to immunotherapy.^29–31^ A recent national qualitative study highlighted the development of effective NRAS-targeted therapies as a key priority for patients.^32^
NRAS-mutated genotype was associated with older adults, cutaneous sites and the limbs, in keeping with current literature.^10,33^ Growing evidence suggests NRAS mutations may be linked with chronic ultraviolet exposure, which may explain these observations.^34,35^ This study found men were more likely to have NRAS mutations, which is less well reported.^10^ In keeping with several meta-analyses, KIT mutations were more prevalent in mucosal melanomas.^10,36^ Survival analysis suggested KIT mutations did not have an impact on survival; however, analysis was limited by lack of follow-up time and sample size. Several phase II trials have observed modest response rates from tyrosine kinase inhibitors in KIT-mutated melanomas; however, the benefit was short lived.^37–39^ A key area of ongoing future research is to identify targeted therapies for KIT-mutated melanomas, particularly mucosal melanomas.^40^ Efforts to develop novel targeted therapies in NRAS-/KIT-mutated melanomas will hopefully deliver actionable biomarkers in addition to BRAF, reinforcing the importance of NRAS/KIT testing in high-risk melanomas.
Immunotherapy may yield a therapeutic avenue in NRAS- and KIT-mutated melanoma. A recent meta-analysis of 1770 patients from 10 studies concluded NRAS-mutated melanoma demonstrated an increased likelihood of response to immunotherapy compared with NRAS WT.^41^ Further research exploring the effectiveness of immunotherapy in KIT-mutated melanoma is required; however, early signals suggest immunotherapy could offer promise in the treatment of KIT-mutated melanoma.^42^ Although this is encouraging, immunotherapy carries significant toxicities and may be contraindicated or poorly tolerated, endorsing the need for novel targeted therapies for NRAS-/KIT-mutated melanomas.
The main strength of this study is the overall size and quality of the data; however, because of low counts, analysis was limited to descriptive statistics for characteristics by KIT genotype. Additionally, regional differences in the completeness of molecular data submissions may have confounded regional comparisons. The study highlighted that factors such as gender, site, stage and geographical region influenced whether a melanoma was NRAS or KIT tested. As characteristics and survival by genotype were explored in the molecular tested cohort only, this cohort may not be representative of the wider melanoma population who did not undergo testing. This limited generalizability may apply in particular to patients with stage I/II disease who are less likely to be tested. As a result of the low numbers, analyses by ethnic group were restricted to White and a minority ethnic grouping (see Table S2). More diverse ethnicity data could be sourced from other national registries or more years of data.
In conclusion, by utilizing a national dataset, this study increases our understanding of melanoma biomarkers NRAS and KIT. Future endeavours should focus on addressing testing disparities and advancing targeted treatments to improve outcomes for patients with high-risk melanoma. Consistent mutational testing in advanced melanoma will prove imperative to support the development and deployment of targeted therapies. This evidence provides a strong foundation for optimizing melanoma care and contributing to global advancements in precision oncology.
Supplementary Material
llaf543_Supplementary_Data
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1National Institute for Health and Care Excellence . Melanoma: assessment and management. Available at: https://www.nice.org.uk/guidance/ng 14/resources/melanoma-assessment-and-management-pdf-1837271430853 (last accessed 18 December 2025).
- 2Seth R, Agarwala SS, Messersmith H et al Systemic therapy for melanoma: ASCO guideline update. J Cin Oncol 2023; 41:4794–820.10.1200/JCO.23.0113637579248 · doi ↗ · pubmed ↗
- 3Garbe C, Amaral T, Peris K et al European consensus-based interdisciplinary guideline for melanoma. Part 1: diagnostics: update 2022. Eur J Cancer 2022; 170:236–55.35570085 10.1016/j.ejca.2022.03.008 · doi ↗ · pubmed ↗
- 4Michielin O, Van Akkooi A, Ascierto P et al Cutaneous melanoma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2019; 30:1884–901.31566661 10.1093/annonc/mdz 411 · doi ↗ · pubmed ↗
- 5Gonzalez D, Fearfield L, Nathan P et al BRAF mutation testing algorithm for vemurafenib treatment in melanoma: recommendations from an expert panel. Br J Dermatol 2013; 168:700–7.23360189 10.1111/bjd.12248 · doi ↗ · pubmed ↗
- 6Robert C, Grob JJ, Stroyakovskiy D et al Five-year outcomes with dabrafenib plus trametinib in metastatic melanoma. N Eng J Med 2019; 381:626–36.10.1056/NEJ Moa 190405931166680 · doi ↗ · pubmed ↗
- 7Ascierto PA, Casula M, Bulgarelli J et al Sequential immunotherapy and targeted therapy for metastatic BRAF V 600 mutated melanoma: 4-year survival and biomarkers evaluation from the phase II SECOMBIT trial. Nat Comm 2024; 15:146.10.1038/s 41467-023-44475-6PMC 1076167138167503 · doi ↗ · pubmed ↗
- 8Atkins MB, Lee SJ, Chmielowski B et al Combination dabrafenib and trametinib versus combination nivolumab and ipilimumab for patients with advanced BRAF-mutant melanoma: the DREA Mseq trial—ECOG-ACRIN EA 6134. J Clin Oncol 2023; 41:186–97.36166727 10.1200/JCO.22.01763 PMC 9839305 · doi ↗ · pubmed ↗