Study Results Demonstrate Population-Based Effects of SCS on Melanoma Mortality Minimal, Not Significant

The number of melanoma diagnoses, which is 1 of the 5 most common cancers in Germany, has increased by nearly 5 times since 1970. Additionally, malignant melanomas account for approximately 10.3% of all new skin cancer diagnoses. In an attempt to reduce skin cancer-related mortality and morbidity, nationwide skin cancer screening (SCS) was introduced in in 2008; however, it is unclear how participation in screening is related to mortality, or if there is an association between them. A study published in JDDG evaluates the relationship between early-stage melanoma incidence found through SCS and mortality in subsequent years.

The study used data on melanoma incidence from 2005 to 2016 and mortality from 2005 to 2018 from 244 counties in Germany. Patients with incident melanoma from 2005 to 2016, except for cases based on a death certificate only, and all melanoma-related deaths from 2005 to 2018 were included in the analysis. Noninvasive (in situ) and stage T1 melanomas (≤ 1 mm tumor thickness) were defined as early stages.

Both high absolute incidences and large increases in incidence could indicate a high intensity of screening and early detection. In addition, the absolute incidence in terms of early-stage age-standardization rates (ASRs) if incidence directly after the introduction of SCS was used as surrogate variables for screening and early detection intensity. Alternatively, the increase in incidence as a result of SCS was calculated as the difference in ASR between the most recent period (2014 to 2016) and the pre-screening period (2005 to 2007).

According to the investigators, an effective SCS can impact mortality either by resulting in a low absolute mortality rate, or by a decrease in mortality. Mortality ASRs were calculated for the 3 most recent 3-year periods (2010 to 2012, 2013 to 2015, and 2016 to 2018) to assess the size of mortality with different delays after SCS introduction. Further, the decrease in mortality was estimated by annual percent change (APC), and the long-term decrease immortality was measured by the different in ASR between the most current time period (2016 to 2018) and the pre-screening period (2005 to 2007), as well as by the APC from 2010 to 2018.

For the population size, the median age at diagnosis and the absolute number of melanomas by T stage at diagnosis are presented descriptively for both the pre-screening period (2005 to 2007) and screening period (2008 to 2010). Further, the relationship between incidence and mortality measures was explored in 2 different ways: Pearson’s correlation coefficients were calculated for each combination of incidence and mortality measure and were presented in a correlation, and linear regressions were fitted. In addition, all models in which the independent variable was the change in early-stage incidence in percent had centered at 100%.

According to the results, the annual number of all new early stages increased by approximately 69% between the pre-screening and screening periods, with an increase of 63% for T1 melanoma and 81% for in situ melanoma. In addition, the median age at diagnosis increased by 7 months. All regressions indicated a lower ASR of mortality with a greater short-term increase of early-stage melanoma incidence.

Further, the correlations between absolute early-stage incidence and mortality vary closely around 0, with the most significant finding being a weak correlation between a larger absolute incidence directly SCS introduction and a stronger mortality decrease in the following 3-year period. Similarly, a stronger short-term incidence change at the time of SCS introduction was weakly correlated with lower absolute mortality variables. Prior research found did not find a significant correlation between absolute incidence and mortality for melanoma, with an increased incidence after the introduction of the SCS being associated with a negative correlation with mortality.

Although not significant, the largest effect in the current regression analyses found between the screening-related incidence increase of early melanomas (APC in 2007 to 2009, introduction of SCS) and the absolute mortality ASR 6 years later. For each additional point of change in the APC of early-stage incidence, the ASR of mortality in 2013 to 2015 was estimated to decreased by 0.0029 melanoma-related deaths per 100,000 inhabitants. The change in early-stage indigence would then correspond to a reduction in melanoma mortality of approximately 2.1%.

According to the investigators, prior research had reported larger effects than those found in the current study. For example, a prior study determined that screening for melanoma can result in a relevant and significant reduction in mortality. Additionally, this study demonstrated a nearly 50% reduction in melanoma mortality by screening in the pilot project of SCS; however, the study is susceptible to biases such as “healthy screenee” bias or overdiagnosis.

Potential factors contributing to the low association in the current study is the low participation rate in SCS (approximately 30% in 2 years), which could be the result of the potentially low relevant and significant population-related effects. In addition, screening participants may be more aware of their health and therefore healthier than nonparticipants, resulting in minimal or no benefit. Further, many of the additional early-stage melanomas that were diagnosed at screening could be overdiagnoses, therefore having no impact on mortality. Screening examinations could also have been carried out at a low quality—in terms of sensitivity—to have an impact on population-based mortality.

Due to the minimal and insignificant effects of SCS on melanoma mortality were estimated, a potential efficacy of screening cannot be demonstrated, according to the investigators. Additional research is needed to further explore or confirm these findings.

Reference

Schumann, L, Eisemann, N, Augustin, J, et al. Association of early-stage incidence and mortality in malignant melanoma – a population-based ecological study. JDDG: Journal der Deutschen Dermatologischen Gesellschaft. 2023; 21(suppl_5):33-40. doi:10.1111/ddg.15218