Melanoma mortality dropped significantly in one German state after a population-wide screening project, while mortality rates remained stable in the rest of Germany.
The effects of organized skin cancer screening on diagnosis and rates of morbidity and mortality are unknown. A prospective, national, randomized trial would be prohibitively expensive, and smaller studies lack sufficient power to demonstrate significance. In Germany, a monumental compromise allowed these authors to evaluate the feasibility of systematic population-based skin cancer screening.
The SCREEN (Skin Cancer Research to Provide Evidence for Effectiveness of Screening in Northern Germany) project was carried out between July 2003 and June 2004 in Schleswig-Holstein, a northern German state with a population of 2.8 million. Screening was offered to state-insured residents of Schleswig-Holstein aged 20 years or older who were not already receiving care for skin cancer. Overall, 1.88 million residents met these criteria, 360,288 of whom were examined during the 1-year screening period. In all, 98% of dermatologists and almost two thirds of nondermatologists took part in SCREEN training. Screening took place via one of two pathways, chosen by the participants:
Pathway A (chosen by 77% of participants) involved an initial evaluation by a nondermatologist trained to screen for and recognize skin cancers. Of these participants, 26% had suspicious lesions or were otherwise deemed to be at risk because of personal or family history and were referred to a dermatologist for follow-up; 37% of such referrals did not reach a dermatologist.
Pathway B involved initial screening by a dermatologist; 57% of these participants were considered at-risk. In all, 81,032 people were considered at-risk, and 20% of at-risk individuals had biopsies. Of the 15,983 biopsies, 18% proved to be skin cancers, including 585 melanomas or lentigo malignas. The authors estimate that nearly half of all melanomas registered during the study period in Schleswig-Holstein were detected through this screening process.
Although this was not a randomized trial, investigators compared melanoma incidence prior to and during the SCREEN period as a temporal control and in Schleswig-Holstein versus other German states as a geographic control. In Schleswig-Holstein, the incidence of invasive melanoma was 16% higher for men and 38% higher for women during the SCREEN project than in the 2 years before SCREEN, while incidence rates did not change in other parts of Germany. Incidence levels returned to pre-SCREEN levels in Schleswig-Holstein after 2004, when the project ended. Tumor thickness was less than 1 mm in almost 90% of invasive metastatic melanomas detected in Schleswig-Holstein during the study period, and the proportion of small tumors increased from 52% before to 64% during the project period, which the authors ascribe to earlier detection with screening. A statistically significant drop in melanoma mortality occurred in Schleswig-Holstein in the 5 years after the SCREEN project, while mortality rates in the rest of Germany remained stable.
Comment: This extraordinary population-based analysis probably represents the most robust study of skin cancer screening that will ever be performed. Although a randomized trial was not possible, comparisons across time and geography provided a reasonable surrogate analysis. The authors’ core message was that population-based screening detected earlier-stage tumors (in part, reflecting a transient detection bias as prevalent cancers are identified). Reduced melanoma-associated mortality rates occurred in the years after the screening period. It would be nice to take the leap of faith and suggest that screening led directly to reduced mortality; however, the lack of a prospective, controlled design prohibits a firm interpretation. Based on the Schleswig-Holstein experience, 45 million people in Germany aged 35 years and older are currently eligible for a skin cancer screening every 2 years. The results of such a large effort will undoubtedly fuel many Journal Watch comments in the future — Hensin Tsao, MD, PhD
Published in Journal Watch Dermatology December 22, 2011