top of page
    • 3 min read

Skin cancer screening more likely to find melanoma at earlier stage


Skin cancer screening more likely to find melanoma at earlier stage

Clinical Pearls

  • This observational quality improvement study assessed how academic and community-based primary care driven melanoma screenings affected melanoma detection.

  • Among 144,851/595,799 patients ≥35 years-old that were screened at least once over a 4 year period, they were more likely to be diagnosed with in situ (incidence, 30.4 vs 14.4; hazard ratio [HR], 2.6; 95% CI, 2.1-3.1; P < .001) or thin invasive (≤1 mm) melanoma (incidence, 24.5 vs 16.1; HR, 1.8; 95% CI, 1.5-2.2; P < .001) and an additional in situ or thin CM within 60 days of initial screening than their age/sex/race-matched unscreened counterparts.

  • There was also a decreased incidence (though not statistically significant) of thick (>4mm) in screened versus unscreened patients.


Discussion:

Cutaneous melanoma is the 3rd most common type of skin cancer and has an exceptionally high mortality. Estimates for 2022 suggest melanoma-specific mortality will reach 7650 (up from ~7180 in 2021). Efforts to ameliorate these statistics include improvements in secondary and tertiary prevention of CM in the way of new treatment modalities and but also increased screening (as seen with other common malignancies).


This 4-year single-institution observational study analyzed stratified Breslow depths of patients that opted to have skin cancer screening (versus those who did not) by primary care physicians, physician’s assistant, nurse practitioners who received ~1 year of web-based training on identifying potentially malignant (i.e., cutaneous melanomas, non-melanoma skin cancers) and benign (i.e., seborrheic keratoses, cherry angiomas) during full body skin exams.


Overall 595,799 patients were included in the analysis (144,851 screened v. 450,948 unscreened). Over 4 years, 246/811 melanomas were identified among the screened participants. After age/sex/race-adjustment, screened patients were more likely to be diagnosed with CM (Hazard Ratio 1.8, 95% confidence interval 1.6-2.1, P<.001), especially in situ (HR 2.6, 95%CI 2.1-3.1, P<.001) and CM <1mm (HR 1.8, 95%CI 1.5-2.2, P<.001) and interval melanomas (in situ HR 2.1, 95%CI 1.7-2.6, P<.001; CM <1mm, HR 1.3, 95%CI 1.0-1.7, P=.03). While there was a trend towards thicker CM (>2 mm, >4mm) in unscreened populations, it was not statistically significant.


The greatest limitation of the study is the use of Breslow thickness (BT) as an endpoint as opposed to American Joint Committee on Cancer 8th edition (AJCC8), satellitosis/distant-metastasis, or melanoma-specific survival (MSS). While BT has traditionally been a proxy for patient outcomes, it does not account for the fact that 1) by absolute number diagnosed, thinner melanomas comprise a large proportion of both melanomas and melanoma mortality and 2) that thinness is not a perfect correlate for “benign” (i.e., there are likely thin but more aggressive melanomas that, when caught early, can significantly reduce morbidity and morality). Additionally, this study was limited to 4 year follow-up. A similar prospective observational cohort that used patient mortality as an endpoint for colorectal cancer[1] had a follow-up period of multiple decades for several thousand participants. Longer longitudinal analysis may allow for patient-outcome based analysis and account for “sojourn time”[2] wherein a malignancy may be pre/subclinical.


Of note, the screened and unscreened population did have notable demographic differences, with the screened population having a significantly older population with a larger proportion of female and non-Hispanic white patients. This is may be due to patient’s own knowledge of risk of skin cancer and deciding to “opt-in” to the screening as a result. In similar prospective observational studies, general population controls were used as a comparison.


Further limitations to the study include multiple settings but only a single-site limiting generalizability based on study population as well as screenings provided by non-physician providers which may limit the clinical acumen of full body skin exams.


Ultimately this study provided insight into the strength of full body skin exams from non-dermatologist physicians and non-physician providers in a single population. While the study concludes that the increase in thin and in situ CM is a sign of overdiagnosis, it is not possible to know this without either true patient outcomes or additional testing of tumor biology to determine how aggressive the tumor might have been. Further longitudinal cohort studies are long overdue to determine how best to implement full body skin exams in terms of frequency and high-risk patient populations.



References




Comments

bottom of page