Skin CancerAdvanced · 9 min read

High-Risk and Poorly-Differentiated Cutaneous SCC

The aggressive end of cutaneous SCC, defined by NCCN 2025 risk criteria, recognized dermoscopically by red-dominant polymorphous vessels and bleeding, and now actionable through molecular risk stratification (40-GEP) and immunotherapy.

By Dr. Yehonatan KaplanPublished Updated

In brief

High-risk and poorly-differentiated cutaneous SCC accounts for the small but disproportionately morbid fraction of keratinocyte cancers responsible for nearly all SCC-attributable deaths. The 2018-2025 era brought three major advances: the AJCC 8th edition and the alternative Brigham and Women's Hospital (BWH) staging systems for refined anatomic risk; the validation and clinical adoption of the 40-gene expression profile test (DecisionDx-SCC) for biological risk stratification; and the approval and rapid uptake of PD-1 inhibitors (cemiplimab, pembrolizumab) for advanced disease, with the Phase III C-POST trial establishing adjuvant cemiplimab as a new standard for high-risk patients. Dermoscopically, poorly-differentiated SCC reverses the white-dominant pattern of well-differentiated tumors: red color predominates, vessels are diffuse polymorphous, bleeding is common, and keratin features are scarce or absent. NCCN 2025 risk stratification (low, high, very-high-risk) drives surgical, radiation, and systemic treatment decisions.

Clinical content

01Dermoscopic signature of poorly-differentiated SCC (Lallas BJD 2015, n=143): predominant red color (OR 13.3 for poor differentiation), diffuse polymorphous vessels in over 50 percent of the lesion surface (OR 30-120), small-caliber vessels, absence of keratin and white structures, frequent bleeding (OR 11.7), and a flat clinical morphology (OR 4.25). These features inform pre-biopsy risk stratification.

02NCCN 2025 risk criteria define low, high, and very-high-risk groups. High-risk features include: trunk/extremity tumor over 2 cm but under 4 cm or any size on head/neck/hands/feet/pretibia/anogenital; poorly defined borders; recurrent tumor; immunosuppression; site of prior radiation or chronic inflammation; rapid growth; neurologic symptoms; histologic subtype (acantholytic, adenosquamous, metaplastic, desmoplastic); 2-6 mm depth or beyond subcutaneous fat; or perineural invasion. Very-high-risk: tumor over 4 cm any location; poorly differentiated; depth over 6 mm or beyond fat; perineural invasion of nerves >=0.1 mm caliber or deeper than dermis; lymphovascular invasion.

03BWH staging is increasingly preferred for clinical decision-making because it concentrates risk in a smaller, more discriminative high-risk group than AJCC8. BWH features: tumor diameter >=2 cm, poorly differentiated histology, perineural invasion of nerve >=0.1 mm caliber, invasion beyond subcutaneous fat. T1 = 0 features; T2a = 1; T2b = 2-3; T3 = 4 or any bone invasion. The Karia/Schmults and Ruiz publications from Brigham and Women's Hospital established BWH T2b/T3 as identifying patients with substantially elevated nodal metastasis (about 30 percent) and disease-specific death (about 10-20 percent).

04AJCC 8th edition (2018) upstages a tumor to T3 with any single high-risk feature: diameter >=4 cm, depth >6 mm, PNI of nerve >=0.1 mm or deeper than dermis or named nerve, or minor bone erosion. T4a = gross cortical bone invasion; T4b = skull base invasion. Pathologic differentiation was dropped from AJCC8 due to inter-observer variability; this is a key divergence from BWH.

05The 40-gene expression profile test (DecisionDx-SCC, Castle Biosciences) classifies tumors into Class 1 (low metastatic risk, 3-year MFS 94-97 percent), Class 2A (higher risk, 3-year MFS 81-89 percent), and Class 2B (highest risk, 3-year MFS 57-69 percent). It re-stratifies patients within NCCN/BWH high-risk categories, identifying biologically aggressive tumors that would not otherwise be flagged. The Wysong et al (Dermatol Ther 2024) cohort of 897 patients confirmed independent prognostic value beyond clinicopathologic staging.

06Adjuvant radiation therapy benefit prediction (Arron IJROBP 2024): in 920 matched patients, ART benefit was concentrated in 40-GEP Class 2B tumors (5-fold prolongation of metastasis-free interval, 50 percent reduction in 5-year disease progression). Class 1 and Class 2A patients did not show similar ART-specific metastatic benefit. This suggests 40-GEP can guide which patients should be most strongly considered for ART beyond traditional indications (margin positivity, extensive PNI, named nerve involvement, recurrent tumor, immunocompromise).

07Cemiplimab (anti-PD-1, EMPOWER-CSCC 1 trial) became the first systemic agent approved for advanced cSCC in 2018. ORR 47-54 percent, complete response 20 percent, 24-month OS 73 percent. Pembrolizumab (KEYNOTE-629) approved in 2020 with similar activity. PD-L1 status does not predict response in cSCC because the tumor mutational burden is among the highest of any cancer (UV-driven), making cSCC inherently immunogenic.

08Neoadjuvant cemiplimab achieves pathologic complete response in approximately 50 percent and major pathologic response in 60-70 percent of resectable high-risk cSCC (Gross NEJM 2022, multiple Phase II trials). NCCN 2025 endorses neoadjuvant immunotherapy for very-high-risk tumors that are borderline resectable, have rapid growth, in-transit metastasis, lymphovascular invasion, or where surgery alone may not be curative or would cause significant functional impairment.

09The Phase III C-POST trial (results 2025) randomized post-surgery, post-RT high-risk patients to adjuvant cemiplimab vs placebo. Cemiplimab reduced recurrence/death by 68 percent (HR 0.32, p<0.001), 24-month DFS 87 vs 64 percent. This establishes adjuvant cemiplimab as a new standard of care for high-risk cSCC after surgery and radiation, fundamentally changing the post-resection algorithm.

10Mohs micrographic surgery is preferred for high-risk and very-high-risk cSCC because it provides 100 percent margin assessment with tissue conservation. NCCN 2025 strengthened the preference for Mohs/PDEMA in very-high-risk cases. Sentinel lymph node biopsy is considered for recurrent or multi-feature high-risk tumors with normal nodal exam, but its prognostic and therapeutic role remains debated; positive SLNB does not consistently improve outcomes when followed by completion lymphadenectomy or RT.

11Solid organ transplant recipients have 65-250 fold increased SCC risk, more aggressive course, and traditional tumor-centric risk factors (size, depth) lose prognostic significance because immunosurveillance has been removed. The biological aggressiveness selected for in immunocompetent patients is not required in SOTRs. In SOTRs, even small tumors can metastasize. Dose-reduction or switch to mTOR inhibitors (sirolimus, everolimus) reduces new SCC development. The CONTRAC-1 trial established a cemiplimab protocol for kidney transplant recipients (mTOR conversion plus pulsed corticosteroids around each infusion) with ORR 46-50 percent and no allograft rejection.

Key dermoscopic features

Predominant red color
Strongest single dermoscopic predictor of poor differentiation. OR 13.3 in Lallas 2015 multivariate analysis.Red color in over 50 percent of lesion surface, often homogeneous; absence of keratin or white structures.
Diffuse polymorphous vessels
Vessels in over 50 percent of the lesion surface, polymorphic morphology (dotted plus linear-irregular plus hairpin), diffuse rather than peripheral. OR 30-120 for poor differentiation.Mix of vessel types (dotted, linear-irregular, hairpin, sometimes arborizing) distributed across the lesion.
Small-caliber vessels
Small vessel diameter dominates in poorly differentiated tumors (OR 3.16). Large arborizing vessels are unusual in well-differentiated SCC and, when present, suggest poor differentiation or BCC.Fine red vascular structures, often densely packed.
Bleeding within lesion
Frequent in poorly differentiated SCC (OR 11.67). Reflects fragile angiogenesis and tissue destruction.Red-brown hemorrhagic areas covering >10 percent of the lesion surface.
Ulceration
Common in advanced SCC. Yellowish structureless amorphous areas without recognizable vessels or keratin.Yellow-pink crusted erosion with surrounding red rim.
Flat clinical morphology
Poorly differentiated SCCs are 4-fold more likely to be clinically flat than nodular (Lallas 2015). The exophytic well-differentiated nodule is replaced by an indurated red plaque.Flat or slightly elevated red plaque, sometimes with central erosion, on photodamaged or chronically inflamed skin.
Absence of keratin
White/yellow scale or keratin reduces odds of poor differentiation by 97 percent (OR 0.06 for poor differentiation).No central keratin mass, no white circles, no white perivascular halos.
Perineural invasion clinical clues
Pain, paresthesia, dysesthesia, or motor weakness in the distribution of a sensory or motor nerve adjacent to the tumor. Also clinical/radiographic evidence of named nerve involvement.Not directly visualizable on dermoscopy; suggested by lesion location near major facial nerves, large tumor size, recurrent tumor, or symptomatic neuropathy.
Indurated, infiltrative borders
Clinical sign of dermal and subcutaneous invasion. Indurated firm tumor with poorly defined edges.Palpable induration extending beyond visible tumor margin; poorly defined borders dermoscopically (Lallas 2015 risk factor).
Desmoplastic SCC pattern
Subtle dermoscopic appearance: pale firm plaque with fine telangiectasias, may resemble a scar. Histologically dense desmoplastic stroma, infiltrative growth, high PNI rate.Pale pink to skin-colored firm plaque, sparse fine vessels; clinical induration disproportionate to dermoscopic findings should raise suspicion.

High yield clinical points15 pearls in 3 groups

Recognition & pattern analysis

6 points
OR 13.3
Red dominates over white in poorly differentiated SCC. Lallas (BJD 2015): predominant red color OR 13.3, vessels in over 50 percent of surface OR 30-120, bleeding OR 11.67, scale/keratin OR 0.06 for poor differentiation. The flip from white to red is the clearest dermoscopic predictor of aggressive biology.
2
Flat indurated red plaque on actinic skin = poorly differentiated SCC until proven otherwise. Poorly differentiated SCCs are 4-fold more likely to be clinically flat than nodular. Combined with diffuse polymorphous vessels and bleeding, this is the high-risk dermoscopic phenotype.
3
BWH staging is more discriminative than AJCC8 for clinical decisions. BWH T2b/T3 captures most metastases and SCC-deaths in a smaller proportion of the population (about 9 percent) than AJCC8 T2/T3 (about 23 percent), giving better positive predictive value. Use BWH for adjuvant therapy decisions; use AJCC8 for cancer registry communication.
4
Image-guided MRI for suspected named nerve PNI. When pain or numbness in trigeminal or facial nerve distribution accompanies a head/neck SCC, contrast-enhanced MRI is mandatory to assess perineural tumor spread, which alters radiation planning to include the proximal nerve course toward the skull base.
5
SOTR risk factors are different: tumor-centric features lose prognostic value. In immunosuppressed transplant patients, even small tumors metastasize because the immune barrier has been removed. Total tumor count, frequency of development, and patient-level factors (immunosuppression duration, regimen) drive risk more than individual tumor characteristics.
6
Switch SOTRs from calcineurin inhibitors to mTOR inhibitors for SCC prevention. Sirolimus and everolimus reduce new SCC development by approximately 50 percent in renal transplant recipients with prior SCC (Euvrard NEJM 2012, multiple RCTs). Coordinate with transplant team.

Management & treatment

3 points
1
Cemiplimab is first-line for advanced unresectable cSCC. EMPOWER-CSCC 1 ORR 47-54 percent, CR 20 percent, 24-month OS 73 percent. Approved 2018. Use regardless of PD-L1 status because cSCC has very high TMB making it inherently immunogenic.
2
Cemiplimab in kidney transplant recipients: CONTRAC-1 protocol. Cross-taper to mTOR inhibitor pre-cemiplimab plus pulsed corticosteroids around each infusion. ORR 46-50 percent, no allograft rejection in the Phase I trial. Multidisciplinary collaboration with transplant nephrology required.
3
Desmoplastic SCC needs MRI and Mohs. 5-fold higher local recurrence (27 percent) and 7-fold higher metastasis (17 percent) than non-desmoplastic. Subtle clinical appearance (pale firm plaque) and high PNI rate require MRI for staging and complete margin assessment via Mohs.

Recent changes (2022 onward)

6 points
1
40-GEP Class 2B identifies patients who benefit most from adjuvant radiation. Arron (IJROBP 2024) demonstrated that ART produces a 5-fold prolongation in time to metastatic event and approximately 50 percent reduction in 5-year disease progression specifically in Class 2B tumors. Class 1 and 2A tumors do not derive similar metastatic benefit from ART.
2
40-GEP Class 1 identifies low metastatic risk despite high-risk clinicopathologic features. Wysong (Dermatol Ther 2024): in NCCN-defined high-risk patients, Class 1 had 3-year MFS 97 percent, vs Class 2B 56-69 percent. Class 1 supports de-escalation of surveillance and adjuvant therapy.
3
Neoadjuvant cemiplimab achieves pCR in ~50 percent of resectable high-risk cSCC. Gross (NEJM 2022) and subsequent Phase II trials. NCCN 2025 endorses neoadjuvant immunotherapy for borderline resectable, rapidly growing, in-transit metastasis, LVI-positive, or functionally challenging high-risk tumors.
4
Adjuvant cemiplimab is now standard for high-risk cSCC after surgery + RT. C-POST trial: 68 percent reduction in recurrence/death (HR 0.32, p<0.001), 24-month DFS 87 vs 64 percent. Practice-changing in 2025. Discuss with all post-surgery, post-RT high-risk patients.
5
Mohs is preferred for very-high-risk cSCC. 100 percent margin assessment, tissue conservation. 5-year recurrence ~3 percent vs 8-10 percent for standard excision. NCCN 2025 strengthened the Mohs/PDEMA preference for very-high-risk tumors.
6
Perineural invasion definition (NCCN 2025): nerve >=0.1 mm caliber, deeper than dermis, or named nerve involvement clinical/radiographic. Most nerves deep to dermis are >0.1 mm. Symptomatic PNI (pain, paresthesia) is very-high-risk and warrants MRI plus consideration of adjuvant RT covering the proximal nerve course.

Lectures covering this topic1 lecture

Academy 2025ACAD25 · #06
What We Have Learned in the Last Few Years About NMSC
Aimilios Lallas

Notable updates & conceptual milestones6 updates

DecisionDx-SCC 40-GEP test (Castle Biosciences)

2020-2024

Validated in Wysong (J Am Acad Dermatol 2021) and Ibrahim (Future Oncol 2022). Three-tier risk class (1, 2A, 2B) based on 40-gene expression in primary tumor. Class 2B re-identifies high-metastatic-risk patients within NCCN/BWH high-risk categories. Independent prognostic value beyond staging.

40-GEP guides adjuvant radiation benefit prediction (Arron IJROBP 2024)

2024

In 920 matched patients, ART benefit (5-fold prolongation in time to metastatic event) was concentrated in Class 2B tumors. Class 1 and 2A did not show similar metastatic benefit. Provides molecular guidance for ART decisions beyond NCCN clinical criteria.

Cemiplimab approval for advanced cSCC (EMPOWER-CSCC 1)

2018-2020

Migden (NEJM 2018) and Rischin (J Immunother Cancer 2020). Established PD-1 blockade as first-line for unresectable advanced cSCC. ORR 47-54 percent, CR 20 percent, 24-month OS 73 percent.

Neoadjuvant cemiplimab (Gross NEJM 2022)

2022

Pathologic CR ~50 percent, major pathologic response >60 percent in resectable high-risk cSCC. Allows function-preserving surgery for very-high-risk tumors.

C-POST trial: adjuvant cemiplimab (2025)

2025

Phase III: 68 percent reduction in recurrence or death (HR 0.32, p<0.001), 24-month DFS 87 vs 64 percent. Establishes adjuvant cemiplimab as new standard for high-risk cSCC after surgery plus RT.

CONTRAC-1: cemiplimab in kidney transplant recipients

2024

Phase I trial established a safe protocol (mTOR conversion plus pulsed corticosteroids) with ORR 46-50 percent and no allograft rejection. Practice-changing for SOTRs with advanced cSCC who previously had no immunotherapy option.

Bottom line

The aggressive end of cutaneous SCC, defined by NCCN 2025 risk criteria, recognized dermoscopically by red-dominant polymorphous vessels and bleeding, and now actionable through molecular risk stratification (40-GEP) and immunotherapy.

15 clinical points · 6 recent updates · 17 references

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, Pyne J, Kyrgidis A, et al. The clinical and dermoscopic features of invasive cutaneous squamous cell carcinoma depend on the histopathological grade of differentiation. Br J Dermatol. 2015;172(5):1308-1315.
    PubMed: 25363081DOI: 10.1111/bjd.13510· Established that dermoscopic features predict differentiation grade. Predominant red color OR 13.3, vessels in over 50 percent OR 30-120, bleeding OR 11.7, white features OR 0.06 for poor differentiation.
  2. [2]
    National Comprehensive Cancer Network. Squamous Cell Skin Cancer (Version 1.2025). NCCN Clinical Practice Guidelines in Oncology. Published January 17, 2025.
    · Current standard of care document for cSCC management; defines low/high/very-high-risk criteria and integrates neoadjuvant and adjuvant immunotherapy.
  3. [3]
    Karia PS, Jambusaria-Pahlajani A, Harrington DP, Murphy GF, Qureshi AA, Schmults CD. Evaluation of American Joint Committee on Cancer, International Union Against Cancer, and Brigham and Women's Hospital tumor staging for cutaneous squamous cell carcinoma. J Clin Oncol. 2014;32(4):327-334.
    PubMed: 24366933DOI: 10.1200/JCO.2012.48.5326· Foundational paper validating BWH staging as superior to AJCC for predicting nodal metastasis and disease-specific death in cSCC.
  4. [4]
    Ruiz ES, Karia PS, Besaw R, Schmults CD. Performance of the American Joint Committee on Cancer Staging Manual, 8th Edition vs the Brigham and Women's Hospital Tumor Classification System for Cutaneous Squamous Cell Carcinoma. JAMA Dermatol. 2019;155(7):819-825.
    PubMed: 30969315DOI: 10.1001/jamadermatol.2019.0032· Confirmed BWH superiority over AJCC8 for clinical decision-making by concentrating risk in a smaller, more discriminative high-risk group.
  5. [5]
    Wysong A, Newman JG, Covington KR, et al. Validation of a 40-gene expression profile test to predict metastatic risk in localized high-risk cutaneous squamous cell carcinoma. J Am Acad Dermatol. 2021;84(2):361-369.
    PubMed: 32344066DOI: 10.1016/j.jaad.2019.07.075· Initial clinical validation of the DecisionDx-SCC 40-GEP test.
  6. [6]
    Ibrahim SF, Kasprzak JM, Hall MA, et al. Enhanced metastatic risk assessment in cutaneous squamous cell carcinoma with the 40-gene expression profile test. Future Oncol. 2022;18(7):833-847.
    PubMed: 34821148DOI: 10.2217/fon-2021-1277· Extended validation cohort of 420 high-risk cSCC patients confirming independent prognostic value of 40-GEP beyond staging.
  7. [7]
    Arron ST, Canueto J, Siegel J, et al. Association of a 40-gene expression profile with risk of metastatic disease progression of cutaneous squamous cell carcinoma and specification of benefit of adjuvant radiation therapy. Int J Radiat Oncol Biol Phys. 2024;120(5):1369-1380.
    PubMed: 38810706DOI: 10.1016/j.ijrobp.2024.05.022· Pivotal study showing 40-GEP Class 2B identifies the subset benefiting most from ART (5-fold prolongation in time to metastatic event, ~50 percent reduction in 5-year progression). Class 1/2A did not show similar metastatic benefit.
  8. [8]
    Wysong A, Somani AK, Ibrahim SF, et al. Integrating the 40-gene expression profile (40-GEP) test improves metastatic risk-stratification within clinically relevant subgroups of high-risk cutaneous squamous cell carcinoma (cSCC) patients. Dermatol Ther (Heidelb). 2024;14(3):593-612.
    PubMed: 38424384DOI: 10.1007/s13555-024-01111-5· 897-patient cohort confirming 40-GEP independent prognostic value across NCCN/BWH/AJCC8 risk strata, including BWH T1/T2a and elderly subgroups.
  9. [9]
    Migden MR, Rischin D, Schmults CD, et al. PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma. N Engl J Med. 2018;379(4):341-351.
    PubMed: 29863979DOI: 10.1056/NEJMoa1805131· Pivotal trial establishing cemiplimab as first PD-1 inhibitor approved for advanced cSCC. ORR 47 percent in metastatic, 44 percent in locally advanced.
  10. [10]
    Rischin D, Khushalani NI, Schmults CD, et al. Phase 2 study of cemiplimab in patients with metastatic cutaneous squamous cell carcinoma: primary analysis of fixed-dosing, long-term outcome of weight-based dosing. J Immunother Cancer. 2020;8(1):e000775.
    PubMed: 32554615DOI: 10.1136/jitc-2020-000775· Long-term EMPOWER-CSCC 1 follow-up: ORR 54 percent, CR 20 percent, 24-month OS 73 percent.
  11. [11]
    Hughes BGM, Munoz-Couselo E, Mortier L, et al. Pembrolizumab for locally advanced and recurrent/metastatic cutaneous squamous cell carcinoma (KEYNOTE-629 study). Ann Oncol. 2021;32(10):1276-1285.
    PubMed: 34293460DOI: 10.1016/j.annonc.2021.07.008· Established pembrolizumab as second PD-1 inhibitor approved for advanced cSCC; ORR 35-50 percent, durable responses >47 months.
  12. [12]
    Gross ND, Miller DM, Khushalani NI, et al. Neoadjuvant cemiplimab for stage II to IV cutaneous squamous-cell carcinoma. N Engl J Med. 2022;387(17):1557-1568.
    PubMed: 36094839DOI: 10.1056/NEJMoa2209813· Phase II trial of neoadjuvant cemiplimab: pCR 51 percent, MPR 13 percent. Established neoadjuvant PD-1 as a function-preserving option for high-risk resectable cSCC.
  13. [13]
    Hanna GJ, Ruiz ES, LeBoeuf NR, et al. Phase I/II study of cemiplimab for kidney transplant recipients with advanced cutaneous squamous cell carcinoma. J Clin Oncol. 2024;42(10):1021-1029.
    PubMed: 38252908DOI: 10.1200/JCO.23.01498· CONTRAC-1: established a safe protocol for cemiplimab in kidney transplant recipients with mTOR conversion and pulsed corticosteroids. ORR 46-50 percent, no allograft rejection.
  14. [14]
    Rischin D, Porceddu S, Day F, et al. Adjuvant Cemiplimab or Placebo in High-Risk Cutaneous Squamous-Cell Carcinoma. N Engl J Med. 2025;393(8):774-785.
    PubMed: 40454639DOI: 10.1056/NEJMoa2502449· C-POST Phase III: 68% reduction in recurrence or death with adjuvant cemiplimab; 24-month DFS 87% vs 64%; Grade 3+ AE 24% vs 14%.
  15. [15]
    Schmults CD, Karia PS, Carter JB, Han J, Qureshi AA. Factors predictive of recurrence and death from cutaneous squamous cell carcinoma: a 10-year, single-institution cohort study. JAMA Dermatol. 2013;149(5):541-547.
    PubMed: 23677079DOI: 10.1001/jamadermatol.2013.2139· Foundational BWH cohort establishing risk factors for SCC recurrence, metastasis, and death.
  16. [16]
    Cavalieri S, Ottini A, Bergamini C, et al. Adjuvant anti-PD-1 therapy in high-risk cutaneous squamous-cell carcinoma. Immunooncol Technol. 2025;29:101557.
    PubMed: 41567503DOI: 10.1016/j.iotech.2025.101557· C-POST plus KEYNOTE-630 reconstructed individual-patient meta-analysis: combined HR 0.40 in nodal high-risk patients; cross-trial comparisons hypothesis-generating.
  17. [17]
    Petrelli F, Dal Cin E, Carioli D, et al. High-risk cutaneous squamous cell carcinoma: a review of current evidence and emerging standards. J Laryngol Otol. 2026;140(3):302-309.
    PubMed: 41549630DOI: 10.1017/S0022215126104332· 2026 review consolidating high-risk criteria and current systemic therapy landscape; neoadjuvant PD-1 and ctDNA-guided surveillance flagged as next.