Skin CancerCore · 8 min read

Basal Cell Carcinoma: Diagnostic Framework

BCC is the most common human cancer; dermoscopy raises sensitivity to 91 to 99 percent and lets the clinician predict subtype before biopsy.

By Dr. Yehonatan KaplanPublished Updated

In brief

Basal cell carcinoma is the most common malignancy in white populations, and its incidence is rising globally. Treatment depends heavily on histopathologic subtype: superficial lesions can be managed with topical agents, photodynamic therapy, or curettage, while nodular, infiltrative, and morpheaform variants require excision, often with margin control. Dermoscopy raises diagnostic accuracy for BCC into the range of 91 to 99 percent and, when combined with reflectance confocal microscopy or LC-OCT, can predict the histologic subtype before any tissue is taken.

Clinical content

01The Menzies criteria remain the foundational dermoscopic algorithm for pigmented BCC. Diagnosis requires the absence of a pigment network (a negative feature) plus at least one of six positive features: large blue-grey ovoid nests, multiple blue-grey globules, maple leaf-like areas, spoke-wheel areas, ulceration, and arborizing telangiectasias. In Menzies' original work this combination yielded a sensitivity of 97 percent and specificity of 92 to 93 percent for separating pigmented BCC from melanoma and nevi.

02Vessel morphology is the single most useful clue in nonpigmented lesions. Sharp, in-focus, branched arborizing telangiectasias are the dermoscopic hallmark of nodular BCC, reported in roughly 75 percent of nodular tumors and 76 percent of infiltrative tumors. Short fine telangiectasias, smaller and less branched, are the dominant vascular pattern of superficial BCC (around 60 percent of cases) and are essentially never seen as the only vessel type in melanoma. Polymorphous, hairpin, dotted, or glomerular vessels do not exclude BCC: in Reiter's systematic review of 5950 BCCs, 6 percent showed glomerular vessels and 10 percent showed a blue-white veil.

03Color cues track histopathology directly. Brown maple leaf-like, spoke-wheel, and concentric structures correspond to pigmented basaloid nests at the dermoepidermal junction and are characteristic of superficial BCC. Blue-grey ovoid nests and multiple blue-grey globules correspond to larger pigmented basaloid aggregates deeper in the dermis and are characteristic of nodular and other nonsuperficial subtypes. White structures (porcelain-white areas, shiny white blotches, chrysalis-like streaks) reflect dermal fibrosis or fibrotic tumor stroma and become the dominant clue in morpheaform and infiltrative variants.

04Pigmentation patterns split BCC into two diagnostic worlds. Pigmented BCC presents with classic Menzies features and is recognized with greater than 95 percent accuracy. Nonpigmented BCC depends almost entirely on vessels, ulceration, multiple small erosions, and shiny white-red structureless backgrounds. Reiter's meta-analysis confirmed that pigmented structures appear in only 0 to 2 percent of clinically nonpigmented BCC, whereas vascular and shiny white clues remain visible regardless of pigmentation.

05Reflectance confocal microscopy maps each dermoscopic feature to a tissue correlate at near-histologic resolution. Cords of basaloid cells connected to the epidermis correspond to superficial BCC, large bright tumor islands with peripheral palisading and clefting correspond to nodular BCC, and dark silhouettes (hyporefractile imprints embedded in bright collagen) correspond to infiltrative BCC. In Longo's 2014 series the combined RCM features cords-connected-to-epidermis plus big-tumor-islands correctly classified 73 to 74 percent of subtypes.

06LC-OCT and OCT extend imaging deeper than RCM, providing cross-sectional views down to roughly 1 mm. OCT achieves a sensitivity around 87 percent and specificity around 80 percent for superficial BCC, rising to about 87 percent accuracy when combined with dermoscopy. These modalities are especially useful on the face and for margin mapping prior to Mohs.

07Diagnostic accuracy is operator-dependent. Chen's 2024 meta-analysis of 100 studies showed experienced dermatologists achieve 83.7 percent sensitivity and 87.4 percent specificity for keratinocytic carcinomas using dermoscopy, a 2.5-fold improvement over clinical examination alone. Adjunctive handheld RCM in Longo's 2024 prospective study of 1005 lesions raised dermoscopy sensitivity from 93.2 to 97.8 percent and specificity from 51.7 to 86.8 percent, with 92 percent of dermoscopy false negatives correctly reclassified as BCC by RCM.

Key dermoscopic features

Arborizing telangiectasias
Sharp, branched, in-focus large-stem vessels; hallmark of nodular and infiltrative BCCDilated neovascular tumor vessels in superficial dermis, prevalence 75 percent in nodular and 76 percent in infiltrative BCC
Short fine telangiectasias
Short, linear, minimally branched vessels; predominant pattern in superficial BCCTelangiectatic vessels in papillary dermis, prevalence 60 percent in superficial BCC
Large blue-grey ovoid nests
Strongest dermoscopic predictor of nonsuperficial BCC; presence makes superficial diagnosis 22 times less likely in flat lesionsLarge pigmented basaloid aggregates invading the dermis; prevalence 36 percent in nodular BCC
Multiple blue-grey dots and globules
Pigmented basaloid nests in papillary or reticular dermis; not specific for any single subtypeRound to oval well-circumscribed pigmented structures, smaller than nests
Maple leaf-like areas
Strong positive predictor of superficial BCC; absent or rare in nodular and morpheaform variantsMultifocal pigmented tumor nests at the dermoepidermal junction with peripheral bulbous extensions
Spoke-wheel areas
Specific for superficial BCC, often peripheralTumor nests connected to epidermis with finger-like projections meeting at a central darker axis
Concentric structures
Considered an early form of spoke-wheel; supports superficial BCCSmall irregular pigmented globules with darker central area
Ulceration
Loss of epidermis with hematogenous crust; common in nodular, infiltrative, and morpheaform variantsLarge structureless red to black-red areas, prevalence 31 to 58 percent across nonsuperficial subtypes
Multiple small erosions
Strong predictor of superficial BCC; rare in nodular variantsSmall brown-red to brown-yellow crusts overlying superficial epidermal loss; prevalence 43 percent in superficial BCC
Shiny white-red structureless areas
Translucent background reflecting fibrotic stroma; very common in superficial BCC (around 79 percent) and infiltrative BCCDiffuse dermal fibrosis or fibrotic tumor stroma, polarization-dependent
Shiny white blotches and strands (chrysalis pattern)
Polarization-only feature; second most common dermoscopic clue to BCC after vessels (49 percent overall)Orthogonal short crossing white lines corresponding to collagenous stromal fibrosis
Absence of pigment network
Required negative criterion in the Menzies algorithm; differentiates BCC from melanocytic lesionsNo reticular pigmentation; basaloid pigment forms structures other than network

High yield clinical points15 pearls in 5 groups

Recognition & pattern analysis

6 points
1
Sharp branched vessels mean BCC until proven otherwise. Large-stem arborizing telangiectasias that branch into finer capillaries and remain in focus across the lesion are present in 59 to 75 percent of BCCs. Out-of-focus or comma vessels point elsewhere.
2
Multiple small erosions favor superficial BCC over Bowen disease. Small brown-red crusts in a flat pink lesion are 7.8-fold more likely to indicate superficial BCC than other subtypes. Bowen tends to show glomerular or coiled vessels and yellow scale instead.
3
Map dermoscopic colors to depth. Brown structures (leaf-like, spoke-wheel, concentric) reflect pigment at the dermoepidermal junction and indicate superficial disease. Blue-grey nests and globules sit deeper in the dermis and indicate nonsuperficial disease.
4
RCM cords vs islands is the subtype shortcut. Cords of basaloid cells connected to the epidermis indicate superficial BCC (54-fold higher odds in Longo 2014). Big tumor islands with clefting indicate nodular BCC (17-fold higher odds for clefting). Dark silhouettes embedded in bright collagen indicate infiltrative BCC.
5
Mind the polarization mode. Shiny white structures and chrysalis are visible only under polarized light. Some vessels look more red and prominent under nonpolarized contact dermoscopy. Switch modes if a lesion seems empty.
6
Use the head-and-neck vs trunk pretest probability. Superficial BCC predominates on the trunk, nodular BCC on the head and neck. A flat pink lesion on the chest with shiny white-red areas and small erosions is statistically far more likely to be superficial than nodular.

Diagnostic criteria & thresholds

3 points
1
Apply the Menzies algorithm in two steps. First confirm there is no pigment network; then look for at least one of: arborizing vessels, large blue-grey ovoid nests, multiple blue-grey globules, maple leaf-like areas, spoke-wheel areas, or ulceration. Sensitivity 97 percent, specificity 92 to 93 percent for pigmented BCC.
2
Short fine telangiectasias point to superficial BCC. When vessels are short, linear, and minimally branched, the lesion is more likely superficial than nodular. Sensitivity 81.9 percent for sBCC when combined with maple leaf-like areas (Lallas 2013 algorithm).
3
Pigmentation drives accuracy. Dermoscopy diagnostic accuracy is higher for pigmented than nonpigmented BCC because pigmented structures are easy to recognize. For amelanotic lesions vessels and shiny white structures become primary clues.

Pitfalls & mimics

4 points
1
Blue-grey ovoid nests rule out superficial BCC. In Lallas' study of 313 BCCs, ovoid nests reduced the odds of superficial diagnosis 11.9-fold; in flat lesions the negative predictive effect rose to 22-fold. If you see them, do not commit to topical therapy.
2
White is not always reassuring. Shiny white blotches and strands appear in 49 percent of BCCs and are the second most common dermoscopic feature after arborizing vessels. They are polarization-dependent and easy to miss on nonpolarized exam.
3
Do not chase glomerular vessels alone. Glomerular vessels in 6 percent of BCCs and blue-white veil in 10 percent should not steer you away from a BCC diagnosis when other classic features are present.
4
Image quality matters. Press lightly with ultrasound gel to preserve vessels; avoid blanching. Capture at 10x with both polarized and nonpolarized light when possible to extract the full set of features.

When to biopsy

1 point
1
Subtype before treatment, not after. Histopathologic subtype guides therapy: imiquimod or PDT for superficial, excision for nodular, Mohs for infiltrative or morpheaform. Pretreatment subtype prediction with dermoscopy plus RCM avoids both biopsy and inappropriate topical therapy.

Recent changes (2022 onward)

1 point
1
Adjunctive RCM rescues amelanotic false negatives. In Longo's 2024 prospective study of 1005 lesions, all 50 dermoscopy false negatives were amelanotic and 46 of these (92 percent) were correctly diagnosed as BCC on handheld RCM.

Lectures covering this topic9 lectures

AAD 2026U003 · #01
Cases Where Confocal Saved the Day
Harold Rabinovitz, MD
AAD 2026U004 · #02
Moving Beyond the Basics: Dermoscopy in Streamlining BCC Management
Elizabeth V. Seiverling, MD, FAAD
Excellence (main)DE-MAIN · #03
Basal Cell Carcinoma and Squamous Cell Carcinoma
G. Argenziano and A. Lallas
Excellence Intl 2025DE-INTL · #01
Decoding My Mind: Non-Pigmented Lesions
Giuseppe Argenziano
Excellence Intl 2025DE-INTL · #11
Diving Into the Blue
Aimilios Lallas
Excellence PracticalDE-PRACT · #02
Facial Cases
Giuseppe Argenziano and Aimilios Lallas
Excellence PracticalDE-PRACT · #05
Difficult Differentials of BCC and SCC
Giuseppe Argenziano and Aimilios Lallas
Academy 2025ACAD25 · #06
What We Have Learned in the Last Few Years About NMSC
Aimilios Lallas
Academy 2025ACAD25 · #14
Pink Tumors and Blood Vessels
Giuseppe Argenziano

Notable updates & conceptual milestones7 updates

Handheld RCM as a real-time adjunct in clinic

2024

Longo's 2024 prospective multicenter study of 1005 clinically suspicious lesions in 3 Italian centers found that adjunctive handheld RCM raised sensitivity from 93.2 to 97.8 percent and specificity from 51.7 to 86.8 percent, with PPV 95.4 percent and NPV 93.5 percent. The handheld probe is faster than the wide-probe device and works on curved facial surfaces.

BCC multivariate prediction score (BCC-I)

2024

From the same 1005-lesion cohort, Longo derived a nomogram-based score combining clinical, dermoscopic, and RCM features (concentric/spoke-wheel structures, dark silhouettes, small and large tumor islands, cord-like structures, clefting, refractile fibrotic collagen, peripheral palisading). AUC 0.95 for personalized BCC probability.

Skin Cancer Diagnostic Accuracy Meta-Analysis

2024

Chen and colleagues (JAMA Dermatology 2024) pooled 100 studies and confirmed that experienced dermatologists using dermoscopy and dermoscopic images had 2.5-fold higher odds of accurate keratinocytic carcinoma diagnosis vs clinical examination. PCPs underperformed dermatologists for skin cancer, with 13.3-fold lower odds for melanoma. The summary metrics now serve as benchmarks for AI and noninvasive diagnostic tools.

Systematic review of 5950 BCCs

2021

Reiter's 2021 meta-analysis remains the largest synthesis of dermoscopic features. Key updates: shiny white structures are the second most common BCC clue (49 percent overall), no single feature is specific for one subtype, and the constellation of features (vessels + erosions + leaf-like for sBCC; vessels + ovoid nests + ulceration for nBCC; vessels + porcelain white for mBCC) drives accuracy.

Two-step RCM algorithm for equivocal lesions

2012

Guitera's two-step RCM method (J Invest Dermatol 2012) uses 710 consecutive equivocal lesions to triage between melanoma and BCC; sensitivity 91 percent and specificity 91 percent for BCC. Forms the basis for current handheld RCM use.

Cemiplimab for HHi-refractory advanced and metastatic BCC

FDA 2021, NCCN 2025

FDA approved 2021 (locally advanced and metastatic BCC after hedgehog inhibitor failure or intolerance) on the basis of ORR 31 to 32 percent (Stratigos Lancet Oncol 2021; Lewis 2024 long-term update). NCCN BCC v1.2025 lists cemiplimab as preferred second-line and as an option for HHi-intolerant patients.

Deep learning meta-analysis of dermoscopy for BCC

2025

Meta-analysis of dermoscopy-based deep learning across 15 studies reports pooled sensitivity 96 percent, specificity 98 percent, AUC 0.99, exceeding dermatologist sensitivity (75 percent). External validation remains the key gap before clinical adoption.

Bottom line

BCC is the most common human cancer; dermoscopy raises sensitivity to 91 to 99 percent and lets the clinician predict subtype before biopsy.

15 clinical points · 7 recent updates · 14 references

Source content

AAD 2026 · U004 · #02

Moving Beyond the Basics: Dermoscopy in Streamlining BCC Management

Elizabeth V. Seiverling, MD, FAAD · University of Connecticut

References

Sources cited in the lecture content or that underpin the clinical points above. Verify with primary sources before practice changes.

  1. [1]
    Lallas A, Tzellos T, Kyrgidis A, et al. Accuracy of dermoscopic criteria for discriminating superficial from other subtypes of basal cell carcinoma. J Am Acad Dermatol. 2014;70(2):303-311.
    PubMed: 24268311DOI: 10.1016/j.jaad.2013.10.003· Foundational diagnostic algorithm: leaf-like + short fine telangiectasia in absence of arborizing vessels, blue-grey ovoid nests, and ulceration predicts sBCC with 81.9 percent sensitivity and 81.8 percent specificity.
  2. [2]
    Longo C, Lallas A, Kyrgidis A, et al. Classifying distinct basal cell carcinoma subtype by means of dermatoscopy and reflectance confocal microscopy. J Am Acad Dermatol. 2014;71(4):716-724.e1.
    PubMed: 24928707DOI: 10.1016/j.jaad.2014.04.067· RCM-dermatoscopy correlation: cords connected to epidermis define sBCC, big tumor islands with clefting define nBCC, dark silhouettes define iBCC.
  3. [3]
    Cameron MC, Lee E, Hibler BP, et al. Basal cell carcinoma: contemporary approaches to diagnosis, treatment, and prevention. J Am Acad Dermatol. 2019;80(2):321-339.
    PubMed: 29782901DOI: 10.1016/j.jaad.2018.02.083· Comprehensive CME review covering diagnosis with dermoscopy and RCM, NCCN risk stratification, and treatment modalities.
  4. [4]
    Reiter O, Mimouni I, Dusza S, Halpern AC, Leshem YA, Marghoob AA. Dermoscopic features of basal cell carcinoma and its subtypes: a systematic review. J Am Acad Dermatol. 2021;85(3):653-664.
    PubMed: 31706938DOI: 10.1016/j.jaad.2019.11.008· Systematic review of 5950 BCCs from 31 studies; pooled prevalence of every dermoscopic feature by subtype.
  5. [5]
    Longo C, Guida S, Mirra M, et al. Dermatoscopy and reflectance confocal microscopy for basal cell carcinoma diagnosis and diagnosis prediction score: a prospective and multicenter study on 1005 lesions. J Am Acad Dermatol. 2024;90(5):994-1001.
    PubMed: 38296197DOI: 10.1016/j.jaad.2024.01.035· Prospective real-world data: dermoscopy plus handheld RCM achieves sensitivity 97.8 percent, specificity 86.8 percent. Nomogram-based BCC-I score with AUC 0.95.
  6. [6]
    Chen JY, Fernandez K, Fadadu RP, et al. Skin cancer diagnosis by lesion, physician, and examination type: a systematic review and meta-analysis. JAMA Dermatol. 2024;160(12):1314-1324.
    PubMed: 39441591DOI: 10.1001/jamadermatol.2024.4382· Meta-analysis of 100 studies: experienced dermatologists with dermoscopy show 2.5-fold higher odds of accurate keratinocytic carcinoma diagnosis vs clinical examination.
  7. [7]
    Menzies SW, Westerhoff K, Rabinovitz H, Kopf AW, McCarthy WH, Katz B. Surface microscopy of pigmented basal cell carcinoma. Arch Dermatol. 2000;136(8):1012-1016.
    PubMed: 10926737DOI: 10.1001/archderm.136.8.1012· Original Menzies algorithm: absence of pigment network plus one of six positive features.
  8. [8]
    Altamura D, Menzies SW, Argenziano G, et al. Dermatoscopy of basal cell carcinoma: morphologic variability of global and local features and accuracy of diagnosis. J Am Acad Dermatol. 2010;62(1):67-75.
    PubMed: 19828209DOI: 10.1016/j.jaad.2009.05.035· Large dermoscopic series quantifying feature prevalence across 609 BCCs in Italy and Australia.
  9. [9]
    Lewis KD, Peris K, Sekulic A, et al. Final analysis of phase II results with cemiplimab in metastatic basal cell carcinoma after hedgehog pathway inhibitors. Ann Oncol. 2024;35(2):221-228.
    PubMed: 38072158DOI: 10.1016/j.annonc.2023.10.123· Cemiplimab 350 mg IV q3w produced ORR 22% with median OS 50 months in metastatic BCC after HHI failure.
  10. [10]
    Stratigos AJ, Sekulic A, Peris K, et al. Phase 2 cemiplimab in locally advanced BCC after hedgehog inhibitor: extended follow-up. J Am Acad Dermatol. 2024;90(2):414-418.
    PubMed: 37839734DOI: 10.1016/j.jaad.2023.08.111· Extended follow-up confirms durable cemiplimab responses anchoring the NCCN advanced-BCC algorithm.
  11. [11]
    Liu H, Shang G, Shan Q. Deep Learning Algorithms in the Diagnosis of Basal Cell Carcinoma Using Dermatoscopy: Systematic Review and Meta-Analysis. J Med Internet Res. 2025;27:e73541.
    PubMed: 41043135DOI: 10.2196/73541· Meta-analysis: pooled DL sensitivity 96%, specificity 98%, AUC 0.99 across 15 dermoscopy studies.
  12. [12]
    Guida S, Spadafora M, Ciardo S, et al. Improving Basal Cell Carcinoma Subtype Diagnosis with Dermoscopy and Reflectance Confocal Microscopy: A Multicenter Prospective Study. Clin Exp Dermatol. 2026.
    PubMed: 41685950DOI: 10.1093/ced/llag037· Subtype-specific patterns across 281 BCCs; handheld RCM raises pre-operative subtype accuracy.
  13. [13]
    Chrysostomidis A, Kostares E, Saramantos A, et al. Clinical Versus Dermoscopic Evaluation of Tumor Margins Prior to Surgical Excision: A Systematic Review. J Clin Med. 2025;14(17):6014.
    PubMed: 40943774DOI: 10.3390/jcm14176014· Dermoscopy-guided margin marking improves complete histologic clearance to 98.7% versus 80-94% with clinical inspection alone (meta-analysis of 900 NMSCs).
  14. [14]
    Witkowski AM, Burshtein J, Christopher M, et al. Clinical Utility of a Digital Dermoscopy Image-Based AI Device. Cancers (Basel). 2024;16(21):3592.
    PubMed: 39518033DOI: 10.3390/cancers16213592· AI-assisted dermoscopy improved dermatologists' sensitivity from 79% to 86% and specificity from 76% to 81%.