Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 9.
Published in final edited form as: J Am Acad Dermatol. 2026 Jun 11;95(4):1167–1169. doi: 10.1016/j.jaad.2026.06.025

Increased Dendritic Cell Density and Altered Melanin Patterns Distinguish Melanoma from Nevi on CRFF-OCT: A Pilot Study Toward Noninvasive Diagnosis

Luísa Polo-Silveira 1, Rozina B Zeidan 1, Stephen W Dusza 1, Ashfaq Marghoob 1, Chih-Shan Jason Chen 1
PMCID: PMC13345679  NIHMSID: NIHMS2186380  PMID: 42276511

Differentiating nevi from melanoma remains challenging. Accurate microscopic evaluation of melanocyte morphology and melanin distribution aids diagnosis; however, histopathology requires biopsy and carries procedure-related morbidity. Dermoscopy and reflectance confocal microscopy (RCM) provide noninvasive bedside insight but are limited by visualization depth (200 μm) and primarily horizontal imaging.1 Cellular-resolution full-field optical coherence tomography (CRFF-OCT) enables deeper penetration (500 μm), vertical imaging analogous to histopathology, and three-dimensional reconstruction.2

This single-center prospective cross-sectional pilot diagnostic study evaluated whether AI-assisted quantitative analysis of CRFF-OCT images can differentiate melanoma from nevi and normal skin using melanocyte- and melanin-related features.

Biopsy-proven melanocytic lesions and adjacent perilesional skin (≤2 mm from lesion margin) were assessed using a CRFF-OCT system (ApolloVue® OCT Image System, model S100; Apollo Medical Optics, Ltd., Taipei, Taiwan). Perilesional skin served as an internal control to establish patient-specific baseline melanin and cellular metrics. Melanin distribution and epidermal dendritic cells (EDC) were analyzed on vertical and en face images using a previously described computer-aided detection platform.2 The algorithm integrated these datasets into a three-dimensional cube, enabling volumetric melanin quantification while minimizing single-layer bias and excluding hyperreflective non-melanin structures. EDC metrics included number, area, average size, density, and spatial distribution. Statistical analyses used the Kolmogorov–Smirnov, t, and Mann–Whitney U tests.

Seventeen Fitzpatrick skin type II patients were included: 11 melanomas (7 invasive, 4 in situ) and 6 nevi (3 dysplastic, 2 dermal, 1 Spitz). Melanomas demonstrated higher EDC density than nevi and controls (p<0.05) (Figure 1). Melanin density was increased in melanomas and nevi versus controls (p<0.05), while mean melanin granule size was larger in melanomas than controls (p<0.05). Inter-granule distance was greatest in nevi, followed by melanomas and controls (p<0.05). A schematic summarizes these features (Figure 2).

Figure 1.

Figure 1.

Open in a new tab

Boxplots of different parameters by diagnostic category.

Figure 2.

Figure 2.

Open in a new tab

Schematic representation of activated melanocytes and melanin distribution in melanoma, nevi, and control skin.

The melanoma group panel shows the highest density of epidermal dendritic cells (EDC) with irregular, elongated dendrites and increased melanin burden, including larger melanin granules. The nevus group panel demonstrates intermediate AM density with more organized dendritic morphology and the greatest inter-granule distance. Control group exhibits low melanin density with small, scattered granules.

Increased EDC density in melanoma aligns with established histopathologic criteria and prior RCM and OCT findings demonstrating increased junctional and pagetoid melanocyte proliferation, as well as reported increased cellular reflectivity and architectural disorganization in malignant lesions.3,4 These results extend prior qualitative observations by providing objective quantitative metrics from volumetric CRFF-OCT data.

Although melanin density was elevated in melanoma and nevi relative to controls, overlap between groups indicates pigment quantity alone is insufficient for reliable discrimination.5 In contrast, larger mean melanin granule size in melanoma and distinct spatial patterns in nevi support a multiparametric approach. Greater inter-granule distance in nevi is consistent with orderly benign nesting, whereas melanoma shows cellular crowding and irregular pigment distribution.1

Limitations include small sample size, single-center design, restriction to Fitzpatrick skin type II, lesion heterogeneity, and lack of direct histopathologic correlation. Although image acquisition required <5 minutes, current AI-assisted analysis may take hours to days, limiting immediate clinical applicability. These findings should therefore be considered preliminary.

In conclusion, AI-assisted CRFF-OCT enables quantitative assessment of melanocyte density and pigment architecture, demonstrating measurable differences between melanoma, nevi, and perilesional skin. These results suggest that quantitative spatial metrics derived from three-dimensional CRFF-OCT may improve discrimination. Larger multicenter studies are needed to validate these findings, assess reproducibility, and determine clinical utility before routine clinical implementation.

Key Takeaway:

AI-assisted CRFF-OCT enables quantitative assessment of melanocyte density and pigment architecture, demonstrating measurable differences between melanoma, nevi, and perilesional skin.

Acknowledgement:

The authors would like to thank Ms. Jean Yen, Ms. Lynn Huang and Dr. Emily Tsai for their technical supports.

Funding sources:

Supported in part by Memorial Sloan Kettering Cancer Center’s NIH/National Cancer Institute Cancer Center support grant P30 CA008748.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflicts of Interest: None declared.

Patient consent: Not applicable.

AI-Assisted Editing: Parts of this manuscript were edited for grammar, clarity, and style using ChatGPT (OpenAI, San Francisco, CA, USA). The authors reviewed and approved all AI-generated suggestions, and the final content is the authors’ own.

References

  • 1.Carrera C, Marghoob AA. Discriminating Nevi from Melanomas: Clues and Pitfalls. Dermatol Clin. 2016. Oct;34(4):395–409. doi: 10.1016/j.det.2016.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lu LY, Chen YT, Chen IL, Shih YC, Liu RT, Lai YJ, Ng CY. Feasibility of High-Cellular-Resolution Full-Field, Artificial-Intelligence-Assisted, Real-Time Optical Coherence Tomography in the Evaluation of Vitiligo: A Prospective Longitudinal Follow-Up Study. Bioengineering (Basel). 2024. Feb 19;11(2):196. doi: 10.3390/bioengineering11020196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Ensslin CJ, Hibler BP, Lee EH, Nehal KS, Busam KJ, Rossi AM. Atypical Melanocytic Proliferations: A Review of the Literature. Dermatol Surg. 2018. Feb;44(2):159–174. doi: 10.1097/DSS.0000000000001367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Gambichler T, Regeniter P, Bechara FG, et al. Characterization of benign and malignant melanocytic skin lesions using optical coherence tomography in vivo. J Am Acad Dermatol. 2007;57(4):629–637. doi: 10.1016/j.jaad.2007.05.029 [DOI] [PubMed] [Google Scholar]
  • 5.Matthews TE, Piletic IR, Selim MA, Simpson MJ, Warren WS. Pump-probe imaging differentiates melanoma from melanocytic nevi. Sci Transl Med. 2011. Feb 23;3(71):71ra15. doi: 10.1126/scitranslmed.3001604. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES