Site Editor
Soo Park, MD
Soo Park, MD
Posted: Thursday, May 23, 2024
An article published in the Journal of Clinical Oncology reported on the use of multivariable prediction tools to forecast the survival outcomes of patients with stage II melanoma receiving adjuvant systemic therapy. Alexander H.R. Varey, MBChB, PhD, FRACS (Plast), FRCS (Plast), of the University of Sydney, Australia, and colleagues examined data extracted from the Melanoma Institute Australia (MIA) database and developed survival prediction models using multivariable Cox regression analyses. Their findings were then compared with Cox models of the American Joint Committee on Cancer 8th edition (AJCC-8) melanoma staging system. The MIA models seemed to be better at predicting the overall and recurrence-free survival than the AJCC-8 staging models, according to the investigators.
“Accurate individualized prognostic estimates of recurrence-free survival and overall survival would allow patients to more accurately weigh the risks and benefits of adjuvant therapy,” stated Dr. Varey and colleagues.
Data from 3,220 patients diagnosed with stage II (clinical or pathologic) melanoma were extracted from the MIA database and included in this study. Survival prediction models were developed using multivariable Cox regression analyses (MIA models) and externally validated twice using data sets from the United States (n = 800) and the Netherlands (n = 8,612). Model performance was then assessed using C-statistics and calibration plots and compared with Cox models based on AJCC-8 staging (stage models).
Overall findings revealed that the discrimination values of the MIA model (C-statistics) for 5-year and 10-year overall survival were 0.71 (95% confidence interval [CI] = 0.69–0.73) and 0.75 (95% CI = 0.73–0.77). These values were significantly higher than the overall survival C-statistics for AJCC-8–based stage models at both 5 (0.62 [95% CI = 0.60–0.64]) and 10 years (0.61 [95% CI = 0.59–0.63]).
Disclosure: For full disclosures of the study authors, visit coi.ascopubs.org.
JNCCN Spotlights
