To the Editor: The COVID-19 pandemic disrupted health care, including melanoma care. We sought to quantify the impact of the pandemic on melanoma stage at diagnosis. Retrospective cohort data were collected at 12 academic centers nationwide with dedicated melanoma clinics. All sites received institutional review board approval. Melanoma cases were categorized as pre-COVID-19 (March 1, 2019-February 29, 2020) or COVID-19 (March 1, 2020-February 28, 2021) based on biopsy dates. Sites enrolled the first 17 cases per month (204 cases per era), although occasionally fewer cases were collected due to patient volume. Multivariable analysis estimated effects on histopathologic eighth edition American Joint Committee on Cancer staging features, adjusting for demographics. The COVID-19 and pre-COVID-19 eras included 1834 and 2062 melanoma cases, respectively. Patient characteristics are summarized in Supplementary Table I, available via Mendeley at https://data.mendeley.com/datasets/rrrmb39nr7/1. In the COVID-19 era, there were relative increases in patient-detected melanomas (54.0% vs 47.8%; P < .001), nodular subtype (12.7% vs 9.8%; P = .005), mean thickness (1.77 vs 1.49 mm; P < .001), positive ulceration (20.0% vs 15.4%; P < .001), and mean mitotic rate (3.03 vs 2.56 mitoses/mm2; P < .001) compared to the pre-COVID-19 era (Supplementary Table II, available via Mendeley at https://data.mendeley.com/datasets/rrrmb39nr7/1). Relative frequency of stage I (67.6% vs 72.7%; P = .001) melanoma was lower in the COVID-19 era but was higher for stage II (18.3% vs 14.8%; P = .006) and stage IV (6.1% vs 4.6%; P = .045) melanomas. There was a relatively lower proportion of stage I melanomas diagnosed during the COVID-19 pandemic for all months except for January 2020 vs 2021 (Supplementary Fig 1, available via Mendeley at https://data.mendeley.com/datasets/rrrmb39nr7/1). There was no difference between time from biopsy to wide local excision during the 2 eras (Supplementary Table II). On multivariable analysis, odds of thickness >2.0-4.0 mm (adjusted odds ratio [aOR], 1.41; 95% CI, 1.07-1.86; P = .006), thickness >4.0 mm (aOR, 1.50; 95% CI, 1.10-2.04; P = .003), ulceration (aOR, 1.40; 95% CI, 1.13-1.73; P < .001), mitoses >1/mm2 (aOR, 1.41; 95% CI, 1.17-1.71; P < .001), nodular subtype (aOR, 1.42; 95% CI, 1.06-1.89; P = .009), and stage II melanoma (aOR, 1.37; 95% CI, 1.10-1.72, P = .002) were increased during the COVID-19 era (Table I).Table IMultivariable-adjusted odds ratios (aORs)∗All outcomes adjusted for age at diagnosis, sex, race, ethnicity, and personal and family history of melanoma. of diagnosis era (COVID-19 vs pre-COVID-19) by melanoma characteristicsaOR∗All outcomes adjusted for age at diagnosis, sex, race, ethnicity, and personal and family history of melanoma. (95% CI)P-value†P-values were computed using a Wald test for the regression coefficients (log odds ratios). P-values and 95% confidence intervals were adjusted using the false discovery rate method to account for multiple comparisons.Melanoma subtype Superficial spreadingRef- Nodular1.42 (1.06-1.89).009 Acral lentiginous1.04 (0.53-2.01).95 Other/unknown1.16 (0.92-1.45).18Thickness (mm) ≤1.0Ref- >1.0-2.01.13 (0.88-1.44).26 >2.0-4.01.41 (1.07-1.86).006 >4.01.50 (1.10-2.04).003 Unknown1.04 (0.34-3.24)1Ulceration‡There was not enough data for “unknown” ulceration status to estimate odds ratios. NoRef- Yes1.40 (1.13-1.73)<.001Biopsy margin status NegativeRef- Positive1.13 (0.89-1.44).38 Unknown0.85 (0.53-1.36).61Mitotic rate (mitoses/mm2) 0Ref- 11.07 (0.82-1.40).81 >11.41 (1.17-1.71)<.001AJCC clinical stage§The eighth edition American Joint Committee on Cancer (AJCC-8) melanoma staging system was used. IRef- II1.37 (1.10-1.72).002 III1.01 (0.69-1.48)1 IV1.44 (0.94-2.23).09 Unknown1.48 (0.73-3.01).28SLNB result NegativeRef- Positive0.89 (0.62-1.26).98 Other/unknown0.35 (0.03-4.62)1 SLNB not performed0.81 (0.68-0.95).008Bold values indicate statistically significant results, defined as P < .05.AJCC, American Joint Committee on Cancer; aOR, adjusted odds ratio; CI, confidence interval; mm, millimeters; ref, reference; SLNB, sentinel lymph node biopsy.∗ All outcomes adjusted for age at diagnosis, sex, race, ethnicity, and personal and family history of melanoma.† P-values were computed using a Wald test for the regression coefficients (log odds ratios). P-values and 95% confidence intervals were adjusted using the false discovery rate method to account for multiple comparisons.‡ There was not enough data for “unknown” ulceration status to estimate odds ratios.§ The eighth edition American Joint Committee on Cancer (AJCC-8) melanoma staging system was used. Open table in a new tab Bold values indicate statistically significant results, defined as P < .05. AJCC, American Joint Committee on Cancer; aOR, adjusted odds ratio; CI, confidence interval; mm, millimeters; ref, reference; SLNB, sentinel lymph node biopsy. In this nationwide multi-institutional study, we observed increased proportions of more advanced melanomas with aggressive features,1Liu W. Dowling J.P. Murray W.K. et al.Rate of growth in melanomas: characteristics and associations of rapidly growing melanomas.Arch Dermatol. 2006; 142https://doi.org/10.1001/archderm.142.12.1551Crossref Scopus (257) Google Scholar similar to smaller cohort studies abroad and in the United States.2Molinier R. Roger A. Genet B. et al.Impact of the French COVID-19 pandemic lockdown on newly diagnosed melanoma delay and severity.J Eur Acad Dermatol Venereol. 2022; 36: e164-e166https://doi.org/10.1111/jdv.17802Crossref PubMed Scopus (7) Google Scholar,3Davis C.H. Ho J. Greco S.H. et al.COVID-19 is affecting the presentation and treatment of melanoma patients in the Northeastern United States.Ann Surg Oncol. 2021; 29: 1629-1635https://doi.org/10.1245/s10434-021-11086-8Crossref PubMed Scopus (5) Google Scholar Our findings, coupled with declining rates of new melanoma cases nationally, suggest that melanoma cases went undiagnosed during the COVID-19 pandemic and subsequently presented at later stages.4London J.W. Fazio-Eynullayeva E. Palchuk M.B. Sankey P. McNair C. Effects of the COVID-19 pandemic on cancer-related patient encounters.JCO Clin Cancer Inform. 2020; 4: 657-665https://doi.org/10.1200/CCI.20.00068Crossref PubMed Scopus (140) Google Scholar As the proportion of internal diagnoses vs referrals and time from biopsy to wide local excision remained stable in our study, it is unlikely that patients simply sought care elsewhere during the pandemic or that procedural delays caused the observed findings. Prior data suggest that asymptomatic melanomas detected incidentally are diagnosed at earlier stages.5Shellenberger R. Nabhan M. Kakaraparthi S. Melanoma screening: a plan for improving early detection.Ann Med. 2016; 48: 142-148https://doi.org/10.3109/07853890.2016.1145795Crossref PubMed Scopus (32) Google Scholar We observed increased rates of patient-identified melanomas and decreased rates of provider-identified melanomas during the COVID-19 pandemic, with more advanced staging at diagnosis, suggesting the importance of screening high-risk individuals. Limitations of our study include generalizability as it only included academic centers and inability to assess melanoma incidence changes due to unknown total case numbers. Future studies are needed to investigate long-term effects of the COVID-19 pandemic on melanoma morbidity, mortality, and health care costs. Our findings emphasize the importance of minimizing melanoma diagnostic delays. None disclosed. This study would not have been possible without the support of many medical students, physicians, and administrative staff members at the various participating academic institutions. We thank everyone for participating in this project and contributing to the study. We would like to thank the Melanoma Research Foundation for their generous support of this study. The University of Arizona Division of Dermatology would like to acknowledge Dr Delaney Stratton for support with the institutional review board and regulatory process. Letter from the Editor: Delays in melanoma diagnosisJournal of the American Academy of DermatologyVol. 87Issue 3PreviewIn this issue of the JAAD, Trepanowski et al report the results of a nationwide cohort study that demonstrated an increased proportion of advanced melanoma related to the COVID-19 pandemic, suggesting that delays in diagnosis contributed to poor outcomes. While we all hope that pandemic-related clinic closures won't have to be repeated, this was just one example of how delays in diagnosis can affect prognosis. Full-Text PDF