Article Text

Original article
CM-Score: a validated scoring system to predict CDKN2A germline mutations in melanoma families from Northern Europe
  1. Thomas P Potjer1,
  2. Hildur Helgadottir2,
  3. Mirjam Leenheer1,
  4. Nienke van der Stoep1,
  5. Nelleke A Gruis3,
  6. Veronica Höiom2,
  7. Håkan Olsson4,
  8. Remco van Doorn3,
  9. Hans F A Vasen5,
  10. Christi J van Asperen1,
  11. Olaf M Dekkers6,
  12. Frederik J Hes1
  13. on behalf of the Dutch Working Group for Clinical Oncogenetics
    1. 1Department of Clinical Genetics, Leiden University Medical Centre, Leiden, The Netherlands
    2. 2Department of Oncology-Pathology, Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden
    3. 3Department of Dermatology, Leiden University Medical Centre, Leiden, The Netherlands
    4. 4Department of Oncology, Lund University and Skåne University Hospital, Lund, Sweden
    5. 5Department of Gastroenterology and Hepatology, Leiden University Medical Centre, Leiden, The Netherlands
    6. 6Department of Clinical Epidemiology, Leiden University Medical Centre, Leiden, The Netherlands
    1. Correspondence to Dr Frederik J Hes, Department of Clinical Genetics, Leiden University Medical Centre, Leiden 2333 ZA, the Netherlands; F.J.Hes{at}


    Background Several factors have been reported that influence the probability of a germline CDKN2A mutation in a melanoma family. Our goal was to create a scoring system to estimate this probability, based on a set of clinical features present in the patient and his or her family.

    Methods Five clinical features and their association with CDKN2A mutations were investigated in a training cohort of 1227 Dutch melanoma families (13.7% with CDKN2A mutation) using multivariate logistic regression. Predefined features included number of family members with melanoma and with multiple primary melanomas, median age at diagnosis and presence of pancreatic cancer or upper airway cancer in a family member. Based on these five features, a scoring system (CDKN2A Mutation(CM)-Score) was developed and subsequently validated in a combined Swedish and Dutch familial melanoma cohort (n=421 families; 9.0% with CDKN2A mutation).

    Results All five features were significantly associated (p<0.05) with a CDKN2A mutation. At a CM-Score of 16 out of 49 possible points, the threshold of 10% mutation probability is approximated (9.9%; 95% CI 9.8 to 10.1). This probability further increased to >90% for families with ≥36 points. A CM-Score under 16 points was associated with a low mutation probability (≤4%). CM-Score performed well in both the training cohort (area under the curve (AUC) 0.89; 95% CI 0.86 to 0.92) and the external validation cohort (AUC 0.94; 95% CI 0.90 to 0.98).

    Conclusion We developed a practical scoring system to predict CDKN2A mutation status among melanoma-prone families. We suggest that CDKN2A analysis should be recommended to families with a CM-Score of ≥16 points.

    • cancer: dermatological
    • cancer: head and neck
    • clinical genetics
    • genetic epidemiology
    • genetic screening/counselling

    Statistics from


    • Contributors Design and conception of study: TPP, HFAV, OMD, FJH. Data collection and assembly: TPP, HH, ML, NvdS, NAG, VH, HO. Data analysis and interpretation: TPP, HH, ML, RvD, HFAV, CJvA, OMD, FJH. Writing of manuscript: TPP, HH, ML, FJH. Critical review and revision of manuscript: all authors. Submission of manuscript: TPP.

    • Funding This work was supported by the Dutch Cancer Society (UL 2015-7511 and UL 2012-5489); The Swedish Cancer Society (CAN 2013/637, CAN 2014/851 and CAN 2015/283); Genomel (LSHC-CT-2006–018702); ERC advanced grant 2011 (291576); The Radiumhemmet Research Funds (144073) and Regional Funds and Hospital Funds in Lund and Stockholm.

    • Competing interests None declared.

    • Patient consent Not required.

    • Ethics approval Leiden University Medical Centre Ethics Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Collaborators Dutch Working Group for Clinical Oncogenetics (participating members): A Wagner, L van der Kolk, M Ausems, Th Van Os, E M Leter, L Spruijt, K van Engelen, L Berger.

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.