To the Editor: COVID-19 significantly impacted health care, including melanoma management.1 Prior data suggest increased proportions of advanced melanoma during the pandemic, but were limited to large academic centers.2 Little is known regarding differential impacts of COVID-19 on melanoma by treatment center type or region. Using the National Cancer Database (NCDB), we examined melanoma presentation and stage at diagnosis by United States (US) region and facility type before and during the COVID-19 pandemic.
We examined 37,019 patients with first histologically confirmed malignant cutaneous melanoma diagnosed before or during the COVID-19 pandemic (2019 vs 2020) across 1084 facilities that reported cases in both years. Given large population sizes, standardized mean differences were calculated to examine differences by diagnosis year (Supplementary Tables I and II, available via Mendeley at https://doi.org/10.17632/fy6b7253dx.1). We performed multivariable logistic regression analyses to assess melanoma presentation and stage at diagnosis by diagnosis year. Stratified analyses by geographic location and facility type were also conducted. Supplementary Appendix provides Supplementary Methods, available via Mendeley at https://doi.org/10.17632/fy6b7253dx.1.
There was a 16% decrease in melanoma diagnoses in 2020 compared to 2019, with increased odds in 2020 of nodular subtype (adjusted odds ratio [aOR] = 1.14, 95% confidence interval [CI] = 1.07-1.22), T4 melanoma (aOR = 1.31, 95% CI = 1.22-1.41), high mitotic rates (aOR = 1.20, 95% CI = 1.15-1.26), ulceration (aOR = 1.16, 95% CI = 1.10-1.23), positive sentinel nodes (aOR = 1.11, 95% CI = 1.03-1.19), and stage IV disease (aOR = 1.30, 95% CI = 1.11-1.53) (Supplementary Table I, available via Mendeley at https://doi.org/10.17632/fy6b7253dx.1). Melanomas undergoing sentinel lymph node biopsy were significantly thicker in 2020 vs 2019 (2.9 mm vs 2.5 mm, standardized mean difference of 0.81) (Supplementary Table III, available via Mendeley at https://doi.org/10.17632/fy6b7253dx.1). When stratified by region, odds of T4 melanoma were increased everywhere, but odds of T3 melanomas were only higher in the Midwest (aOR = 1.22, 95% CI = 1.07-1.39) and Northeast (aOR = 1.25, 95% CI = 1.07-1.45). Odds of stage IV disease in the Northeast (aOR = 1.60, 95% CI = 1.09-2.36) and positive sentinel nodes in the South (aOR = 1.13, 95% CI = 1.00-1.27) were also increased in 2020 (Table I ). When stratified by facility type, academic and comprehensive community cancer programs experienced higher odds of T2-T4 melanomas, while community cancer programs only experienced higher odds of T4 melanomas (aOR = 1.81, 95% CI = 1.32-2.50). Lastly, odds of positive sentinel nodes increased in community (aOR = 1.51, 95% CI = 1.06-2.16) and comprehensive community programs (aOR = 1.19, 95% CI = 1.05-1.36) in 2020 (Table II ). Additional analyses revealed similar findings among subsequent melanomas and all melanomas in 2019 vs 2020, while these findings were not present when examining initial primary melanomas in 2018 vs 2019 (Supplementary Tables VI-XIV, available via Mendeley at https://doi.org/10.17632/fy6b7253dx.1).
Table I.
Multivariable logistic regression analyses of first cutaneous melanoma presentation by year of diagnosis (during COVID-19 in 2020 vs before COVID-19 in 2019) and geographic location
| Characteristics | All |
West |
Midwest |
Northeast |
South |
|---|---|---|---|---|---|
| aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | |
| Histologic subtype | |||||
| Superficial spreading | Reference | Reference | Reference | Reference | Reference |
| Acral lentiginous | 1.05 (0.90, 1.23) | 1.08 (0.75, 1.54) | 1.07 (0.75, 1.51) | 1.06 (0.72, 1.55) | 1.05 (0.81, 1.35) |
| Nodular | 1.14 (1.07, 1.22)∗∗∗ | 1.10 (0.94, 1.28) | 1.13 (1.00, 1.28)∗ | 1.08 (0.92, 1.26) | 1.21 (1.09, 1.35)∗∗∗ |
| NOS/other | 1.02 (0.97, 1.07) | 1.02 (0.92, 1.14) | 0.92 (0.84, 1.01) | 1.01 (0.91, 1.11) | 1.09 (1.01, 1.18)∗ |
| Anatomic site | |||||
| Head and neck | Reference | Reference | Reference | Reference | Reference |
| Trunk | 0.99 (0.93, 1.05) | 0.95 (0.84, 1.09) | 0.94 (0.83, 1.05) | 0.99 (0.86, 1.13) | 1.03 (0.94, 1.14) |
| Upper extremities | 1.00 (0.94, 1.06) | 0.90 (0.79, 1.03) | 0.93 (0.82, 1.05) | 1.06 (0.92, 1.22) | 1.07 (0.97, 1.18) |
| Lower extremities | 1.01 (0.94, 1.08) | 0.98 (0.85, 1.14) | 0.97 (0.85, 1.12) | 1.03 (0.88, 1.21) | 1.04 (0.92, 1.17) |
| Skin, NOS | 0.89 (0.69, 1.16) | 1.17 (0.63, 2.16) | 0.75 (0.42, 1.30) | 0.87 (0.47, 1.60) | 0.88 (0.58, 1.33) |
| Breslow thickness | |||||
| T1 (≤1 mm) | Reference | Reference | Reference | Reference | Reference |
| T2 (>1-2 mm) | 1.14 (1.08, 1.20)∗∗∗ | 1.07 (0.95, 1.21) | 1.14 (1.02, 1.27)∗ | 1.25 (1.10, 1.41)∗∗∗ | 1.12 (1.02, 1.23)∗ |
| T3 (>2-4 mm) | 1.16 (1.09, 1.24)∗∗∗ | 1.15 (1.00, 1.33) | 1.22 (1.07, 1.39)∗∗ | 1.25 (1.07, 1.45)∗∗ | 1.09 (0.98, 1.21) |
| T4 (>4 mm) | 1.31 (1.22, 1.41)∗∗∗ | 1.32 (1.13, 1.55)∗∗∗ | 1.34 (1.16, 1.55)∗∗∗ | 1.47 (1.25, 1.71)∗∗∗ | 1.21 (1.08, 1.35)∗∗ |
| Mitotic rate | |||||
| 0 | Reference | Reference | Reference | Reference | Reference |
| 0-1 | 1.04 (0.98, 1.10) | 1.06 (0.93, 1.21) | 1.12 (1.00, 1.27) | 1.02 (0.90, 1.16) | 0.98 (0.88, 1.08) |
| >1 | 1.20 (1.15, 1.26)∗∗∗ | 1.23 (1.11, 1.38)∗∗∗ | 1.28 (1.16, 1.41)∗∗∗ | 1.22 (1.09, 1.36)∗∗∗ | 1.13 (1.04, 1.23)∗∗ |
| Ulceration | |||||
| No | Reference | Reference | Reference | Reference | Reference |
| Yes | 1.16 (1.10, 1.23)∗∗∗ | 1.15 (1.02, 1.29)∗ | 1.20 (1.08, 1.34)∗∗∗ | 1.23 (1.09, 1.39)∗∗∗ | 1.11 (1.01, 1.21)∗ |
| Sentinel lymph node | |||||
| Negative | Reference | Reference | Reference | Reference | Reference |
| Positive | 1.11 (1.03, 1.19)∗∗ | 1.10 (0.94, 1.29) | 1.07 (0.94, 1.23) | 1.13 (0.95, 1.34) | 1.13 (1.00, 1.27)∗ |
| Stage (AJCC 8th edition) | |||||
| I | Reference | Reference | Reference | Reference | Reference |
| II | 1.13 (1.07, 1.20)∗∗∗ | 1.16 (1.02, 1.32)∗ | 1.16 (1.04, 1.31)∗ | 1.26 (1.11, 1.44)∗∗∗ | 1.03 (0.94, 1.13) |
| III | 1.20 (1.13, 1.28)∗∗∗ | 1.14 (0.98, 1.31) | 1.25 (1.11, 1.42)∗∗∗ | 1.24 (1.07, 1.44)∗∗ | 1.19 (1.07, 1.32)∗∗ |
| IV | 1.30 (1.11, 1.53)∗∗ | 1.41 (0.95, 2.08) | 1.28 (0.93, 1.77) | 1.60 (1.09, 2.36)∗ | 1.15 (0.88, 1.49) |
Multivariate models were adjusted for age, sex, race, Charlson Comorbidity Index, insurance, median household income, distance from facility, and facility type. Histologic subtype, anatomic site, and geographic location were included as additional covariates if they were not a stratification variable or exposure of interest.
∗∗∗P < .001, ∗∗P < .01, and ∗P < .05.
AJCC, American Joint Committee on Cancer; aOR, adjusted odds ratio; CI, confidence interval; mm, millimeter; NOS, not otherwise specified.
Table II.
Multivariable logistic regression analyses of first cutaneous melanoma presentation by year of diagnosis (during COVID-19 in 2020 vs before COVID-19 in 2019) and facility type
| Characteristics | Academic/Research Program |
Integrated Network Cancer Program |
Comprehensive Community Cancer Program |
Community Cancer Program |
|---|---|---|---|---|
| aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | |
| Histologic subtype | ||||
| Superficial spreading | Reference | Reference | Reference | Reference |
| Acral lentiginous | 1.18 (0.95, 1.46) | 0.87 (0.60, 1.28) | 0.84 (0.59, 1.20) | 1.56 (0.66, 3.83) |
| Nodular | 1.15 (1.05, 1.27)∗∗ | 1.26 (1.08, 1.47)∗∗ | 1.11 (0.98, 1.25) | 0.86 (0.64, 1.18) |
| NOS/other | 0.99 (0.93, 1.06) | 1.01 (0.90, 1.12) | 1.06 (0.98, 1.16) | 0.99 (0.78, 1.24) |
| Anatomic site | ||||
| Head and neck | Reference | Reference | Reference | Reference |
| Trunk | 0.98 (0.90, 1.06) | 0.89 (0.78, 1.03) | 1.08 (0.97, 1.20) | 0.82 (0.61, 1.09) |
| Upper extremities | 1.03 (0.95, 1.13) | 0.88 (0.76, 1.01) | 1.02 (0.92, 1.14) | 0.94 (0.69, 1.27) |
| Lower extremities | 0.95 (0.86, 1.04) | 0.94 (0.80, 1.10) | 1.18 (1.04, 1.34)∗ | 0.94 (0.66, 1.34) |
| Skin, NOS | 0.85 (0.57, 1.25) | 0.72 (0.35, 1.41) | 1.03 (0.66, 1.61) | 1.12 (0.35, 3.63) |
| Breslow thickness | ||||
| T1 (≤1 mm) | Reference | Reference | Reference | Reference |
| T2 (>1-2 mm) | 1.11 (1.02, 1.20)∗ | 1.13 (1.00, 1.29) | 1.17 (1.06, 1.29)∗∗ | 1.28 (0.98, 1.69) |
| T3 (>2-4 mm) | 1.14 (1.04, 1.26)∗∗ | 1.02 (0.88, 1.19) | 1.27 (1.13, 1.43)∗∗∗ | 1.20 (0.87, 1.65) |
| T4 (>4 mm) | 1.31 (1.19, 1.45)∗∗∗ | 1.16 (0.98, 1.36) | 1.34 (1.18, 1.52)∗∗∗ | 1.81 (1.32, 2.50)∗∗∗ |
| Mitotic rate | ||||
| 0 | Reference | Reference | Reference | Reference |
| 0-1 | 1.02 (0.94, 1.11) | 1.11 (0.97, 1.27) | 1.04 (0.93, 1.16) | 0.95 (0.71, 1.28) |
| >1 | 1.18 (1.10, 1.26)∗∗∗ | 1.27 (1.13, 1.43)∗∗∗ | 1.17 (1.07, 1.28)∗∗∗ | 1.48 (1.16, 1.88)∗∗ |
| Ulceration | ||||
| No | Reference | Reference | Reference | Reference |
| Yes | 1.14 (1.06, 1.23)∗∗∗ | 1.14 (1.00, 1.29)∗ | 1.16 (1.05, 1.28)∗∗ | 1.50 (1.16, 1.93)∗∗ |
| Sentinel lymph node | ||||
| Negative | Reference | Reference | Reference | Reference |
| Positive | 1.07 (0.96, 1.19) | 1.02 (0.87, 1.20) | 1.19 (1.05, 1.36)∗∗ | 1.51 (1.06, 2.16)∗ |
| Stage (AJCC 8th edition) | ||||
| I | Reference | Reference | Reference | Reference |
| II | 1.10 (1.01, 1.19)∗ | 1.03 (0.90, 1.18) | 1.19 (1.08, 1.33)∗∗∗ | 1.41 (1.08, 1.86)∗ |
| III | 1.13 (1.03, 1.23)∗ | 1.11 (0.96, 1.29) | 1.32 (1.17, 1.49)∗∗∗ | 1.85 (1.35, 2.54)∗∗∗ |
| IV | 1.27 (1.00, 1.61) | 1.33 (0.89, 2.01) | 1.26 (0.95, 1.67) | 1.75 (0.77, 4.06) |
Multivariate models were adjusted for age, sex, race, Charlson Comorbidity Index, insurance, median household income, distance from facility, and geographic location. Histologic subtype and anatomic site were included as additional covariates if they were not an exposure of interest.
∗∗∗P < .001, ∗∗P < .01, and ∗P < .05.
AJCC, American Joint Committee on Cancer; aOR, adjusted odds ratio; CI, confidence interval; mm, millimeter; NOS, not otherwise specified.
We found that the COVID-19 pandemic may have disproportionately impacted certain US regions and health care facilities, including the Northeast and academic and comprehensive community cancer programs. These findings may be due to differences in COVID-19 restrictions, health care policies, and referral patterns.3 , 4 Due to data limitations in NCDB, we were unable to assess incidence changes and could not separately examine diagnoses in January or February 2020 prior to the pandemic. Of note, NCDB comprises >70% of cancer cases in the United States, and our case mix across regions and facility types did not change significantly between 2019 and 2020 (Supplementary Table II, available via Mendeley at https://doi.org/10.17632/fy6b7253dx.1).5 Further research is needed to comprehensively understand long-term impacts of the pandemic on melanoma outcomes.
Conflicts of interest
Mr Kim is a paid consultant at Verve Therapeutics and SeQure Dx, unrelated to this research. Dr Hartman is a scientific officer for Evereden, unrelated to this research. Mr Kim and Dr Hartman's interests were reviewed and are managed by Mass General Brigham in accordance with their conflict-of-interest policies. Dr Behbahani has no conflicts of interest to declare.
Footnotes
Funding sources: Dr Hartman is funded by the Department of Defense and the VA Integrated Service Network 1 (VISN-1).
IRB approval status: IRB approval was not required due to the use of deidentified and publicly available data.
Reprints not available from the authors.
References
- 1.Chang M.S., Leachman S.A., Berry E.G., et al. Changes in melanoma care practices during the COVID-19 pandemic: a multi-institutional cross-sectional survey. Dermatol Online J. 2021;27(4) 13030/qt4sc3k9r0. [PubMed] [Google Scholar]
- 2.Trepanowski N., Chang M.S., Zhou G., et al. Delays in melanoma presentation during the COVID-19 pandemic: a nationwide multi-institutional cohort study. J Am Acad Dermatol. 2022;87(5):1217–1219. doi: 10.1016/j.jaad.2022.06.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zhang X., Warner M.E. COVID-19 policy differences across US States: shutdowns, reopening, and mask mandates. Int J Environ Res Public Health. 2020;17(24):9520. doi: 10.3390/ijerph17249520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hallas L., Hatibie A., Majumdar S., Pyarali M., Hale T. Univ Oxford; 2021. Variation in US States' Responses to COVID-19. [Google Scholar]
- 5.Boffa D.J., Rosen J.E., Mallin K., et al. Using the National Cancer Database for outcomes research: a review. JAMA Oncol. 2017;3(12):1722–1728. doi: 10.1001/jamaoncol.2016.6905. [DOI] [PubMed] [Google Scholar]