Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

A New Online Tool Predicts RFS and OS Probabilities in Patients with Stage II Melanoma

MIA calculator should help inform clinicians on which patients to select for adjuvant therapy
07 Feb 2024

The Melanoma Institute Australia (MIA) models predict recurrence-free survival (RFS) and overall survival (OS) in patients with a primary cutaneous melanoma considerably better than models based on AJCC-8 staging criteria or using the EORTC nomogram. The clinical impact of accurate prognostic estimates for individual patients with both clinical and pathological stage II disease is significant as it could guide access to adjuvant systemic treatment.

With the current absence of biomarkers that predict response to systemic therapy, this would provide a means of selection for clinical trials of patients with high-risk tumour features who are clinically node-negative and have not had a sentinel node biopsy. The nomograms are presented by Dr. Alexander H. R. Varey of the Melanoma Institute Australia and Faculty of Medicine and Health, The University of Sydney, and Department of Plastic & Reconstructive Surgery, Westmead Hospital in Sydney, NSW, Australia, and colleagues in an article published on 5 February 2024 in the JCO.

The authors wrote in the background that the most used method for determining prognosis is based on the AJCC-8 melanoma staging system, stage groups of which were determined using melanoma-specific survival (MSS), while selecting patients for adjuvant treatment is usually informed by RFS estimates. AJCC-8 relies primarily on the primary tumour's Breslow thickness and ulceration status, regional node status, and the presence of any distant metastases. EORTC nomogram provides estimates of both MSS and RFS, but only for sentinel node–negative patients.

However, since many patients diagnosed with melanoma do not undergo sentinel node biopsy, this significantly limits the applicability of the AJCC-8 staging system or the EORTC tool in clinical practice. Furthermore, clinical trials of novel adjuvant drug combinations are now underway, and sentinel node biopsy is no longer mandatory as an eligibility criterion for primary tumours >1 mm thick, whereas it was for previous adjuvant trials. Patients with T3b or T4 melanomas may now receive adjuvant drugs irrespective of their sentinel status.

The patient's age and sex and other primary tumour features may also have important prognostic implications. These include tumour mitotic rate, presence of tumour infiltrating lymphocytes, presence of regression, presence of lymphovascular invasion, melanoma subtype, and body site. The degree to which each of these additional variables may be independently prognostically significant remains to be determined, particularly for outcomes other than MSS.

Improvements in RFS were demonstrated into recent randomised studies for patients with sentinel node negative stage IIB or IIC melanoma receiving adjuvant systemic treatment with pembrolizumab or nivolumab. However, side effects also occurred. Accurate individualised prognostic estimates would allow to weigh the risks and benefits of adjuvant treatment more accurately. The purpose of this study was to develop a multivariable prediction tool for RFS and OS for patients with both clinical and pathologic stage II melanoma.

Data were extracted from the MIA database for 3,220 patients diagnosed with clinical or pathological stage II melanoma. Survival prediction models were developed using multivariable Cox regression analyses (MIA models) and externally validated twice using data sets from the United States and the Netherlands. Each model's performance was assessed using C-statistics and calibration plots and compared with Cox models based on AJCC-8 staging (stage models).

The 5-year and 10-year RFS C-statistics were 0.70 and 0.73 for MIA model versus 0.61 and 0.60 for stage model. For OS, the 5-year and 10-year C-statistics were 0.71 and 0.75 for MIA model compared with 0.62 and 0.61 for stage-model. The MIA models were well calibrated and externally validated.

The authors concluded as these models were robust on external validations, they may be used in everyday practice both with (ideally) and without performing sentinel node biopsy to identify high-risk patients for further management strategies. The free online tool on the MIA website will allow the estimation of risk both for patients who are sentinel node negative and for those who did not have a sentinel node biopsy.


Varey AHR, Li I, El Sharouni M-A, et al. Predicting Recurrence-Free and Overall Survival for Patients With Stage II Melanoma: The MIA Calculator. JCO; Published online 5 February 2024. DOI: https://doi.org/10.1200/JCO.23.01020

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.