Dive deeper

Open any topic for the full explanation, tables, clinical checkpoints, and references.

Full-length video Where can I learn more about low-risk and high-risk skin cancer?
Why is skin-cancer risk a stack rather than a single word?

Direct Answer: Risk Is a Stack, Not a Single Word

Basal cell carcinoma and squamous cell carcinoma are diagnoses. They are not complete treatment plans.

Two patients can have the same diagnosis and need different care. A small, sharply defined primary BCC on the back may be treated very differently from a recurrent infiltrative BCC on the nose. A thin, well-differentiated cSCC on a low-risk site is not the same disease problem as a poorly differentiated cSCC extending beyond fat in an immunosuppressed patient.1–4

The best mental model is a five-part stack:

  1. Tumor: type, size, borders, growth, and recurrence.
  2. Pathology: subtype, differentiation, depth, nerve or vessel involvement.
  3. Location: whether the anatomy leaves room for a standard margin and repair.
  4. Patient: immune status, health, age, treatment tolerance, and goals.
  5. History: prior surgery, radiation, scar, or incomplete treatment.

The formal risk category comes from diagnosis-specific tumor, pathology, location, immune-status, and recurrence criteria. The treatment plan then adds overall health, life expectancy, treatment tolerance, and patient goals.

“Low risk does not mean harmless, and high risk does not mean hopeless. The label tells us how carefully the cancer should be removed and followed, and whether disease-specific staging deserves discussion.”

- Dr. Hocker
How can several risk factors stack together?

Risk Factors Can Stack

One feature can be enough to change a formal category, but real clinical concern often rises when several features travel together.

A 1.5-cm cSCC on the forearm may look modest by size. Add recurrence, poor differentiation, 5-mm depth, immune suppression, and numbness, and the treatment problem changes completely. Conversely, a single risk label may have a different practical effect when every other feature is favorable.

The point is not to add up risk factors like points in a game. It is to understand which one changes which decision:

  • Margin risk: How likely is invisible tumor to extend beyond what can be seen?
  • Recurrence risk: How likely is the cancer to return at the treated site?
  • Metastatic risk: How likely is it to reach lymph nodes or distant sites?
  • Anatomic risk: What function or structure is endangered by tumor or treatment?
  • Treatment risk: What operation, staging, or follow-up intensity is justified?

Those risks overlap, but they are not identical.

How are risk, grade, and stage different?

Risk, Grade, and Stage Are Different

These terms are often blended together, but they answer different questions.

Risk, Grade, and Stage Are Different
TermThe question it answers
Risk categoryHow likely is this tumor to recur or behave aggressively, and how intensive should local treatment be?
Grade/differentiationHow abnormal do the tumor cells look under the microscope?
StageHow far has the cancer progressed in the skin, lymph nodes, or other organs?

For BCC, risk classification mainly guides local treatment because BCC is far more likely to recur or destroy nearby tissue than to metastasize.5,6

For cSCC, risk classification also matters for nodal spread and disease-specific death. Current NCCN guidance uses low-, high-, and very-high-risk treatment categories, while systems such as the Brigham and Women's Hospital (BWH) classification and American Joint Committee on Cancer (AJCC) staging serve prognostic roles.2,7–9

Melanoma uses its own AJCC staging system based on features such as Breslow thickness, ulceration, lymph nodes, and distant disease.10 It should branch immediately to melanoma-specific guidance.

What are the five groups of skin-cancer risk factors?

The Five Groups of Risk Factors

1. Tumor behavior

Risk tends to rise when a tumor:

  • has returned after prior treatment;
  • has poorly defined clinical borders;
  • is growing rapidly;
  • causes pain, numbness, tingling, or weakness;
  • is large for its location;
  • has been incompletely treated before.1,2

Recurrence matters because scar can obscure the edge and because the first treatment did not clear the tumor's full extent.

2. Pathology

The report may identify:

  • an aggressive BCC growth pattern;
  • poor differentiation in cSCC;
  • greater depth or invasion beyond subcutaneous fat;
  • perineural invasion;
  • lymphovascular invasion;
  • an aggressive cSCC subtype;
  • uncertainty that requires more tissue or review.1–3

Pathology and sampling must be interpreted together. A small biopsy can identify the cancer yet miss a deeper or different component elsewhere.13

3. Location

Location matters for two reasons:

  1. some sites are associated with higher recurrence risk or subclinical extension; and
  2. the cost of removing unnecessary normal tissue is greater around the eyelid, nose, lip, ear, hand, foot, and genital skin.

Current 2026 NCCN tables classify the head and neck as high-risk anatomy for BCC and cSCC at any size. The hands, feet, pretibial skin, anogenital region, and other specified sites also receive special treatment in the guidelines.1,2

Do not self-classify from a body map alone; thresholds differ by cancer type. Tumor type, size, pathology, recurrence, and immune status still matter.

Why high-risk location is not a cosmetic label

High-risk sites often combine three problems:

  1. tumor extensions are harder to see against sun-damaged or structurally complex skin;
  2. a recurrence can threaten an eyelid, nose, lip, ear, digit, or another functional structure; and
  3. removing an unnecessarily wide margin can make reconstruction harder.

That is why Mohs is particularly logical on high-stakes anatomy when the diagnosis is appropriate for Mohs. The operation addresses both sides of the problem: find the microscopic edge and avoid sacrificing the same amount of tissue in every direction.

4. Patient factors

Immune suppression can alter skin-cancer biology. Solid-organ transplant recipients, patients with hematologic malignancy, and people taking certain immune-suppressing medicines can develop more aggressive or numerous cSCCs. Current BCC and cSCC guidance treats immune suppression as a risk factor.1,2

Patient health can also move the decision in the other direction. Observation can be reasonable for selected low-risk BCC when severe frailty or limited life expectancy makes treatment more burdensome than the disease. Risk is about the disease and the cost of treatment.12

5. Prior treatment and tissue history

Prior radiation, scar, chronic inflammation, and previous excision can change anatomy and tumor behavior. A recurrent tumor in scarred tissue is harder to judge clinically. Prior radiation can also reduce reconstructive options and healing reserve.1,2

What makes a BCC lower risk or higher risk?

BCC: What Makes It Lower Risk or Higher Risk?

Current NCCN BCC guidance uses low- and high-risk categories. One high-risk feature can move the tumor into the high-risk pathway.1

Features that support a lower-risk BCC pathway

  • primary tumor;
  • clearly defined borders;
  • low-risk site and size;
  • nodular or superficial subtype without an aggressive component;
  • no perineural involvement;
  • no immune suppression or prior radiation at the site.1

For location and size, a primary BCC on the trunk or extremities is low risk by that factor when it is under 2 cm; 2 cm or larger moves it to high risk. Certain head, neck, hand, foot, pretibial, and anogenital sites are high risk at any size under current guidance.1

Features that support a high-risk BCC pathway

  • recurrent tumor;
  • poorly defined borders;
  • high-risk location or size;
  • immune suppression;
  • prior radiation at the site;
  • infiltrative, morpheaform/sclerosing, micronodular, basosquamous, sarcomatoid differentiation, or another aggressive pattern;
  • perineural involvement.1

The main danger of high-risk BCC is usually local: recurrence, deep extension, tissue destruction, and a more difficult reconstruction. Metastatic BCC is rare, but locally destructive disease can still have major consequences.5,6

Why is the cSCC risk framework more serious?

cSCC: Why the Risk Framework Is More Serious

cSCC can recur locally, spread to lymph nodes, and cause death. Current NCCN guidance therefore separates low, high, and very high risk.2

Selected current cSCC thresholds

  • On the trunk or extremities, a tumor under 2 cm is low risk by location/size, a tumor from 2 cm through exactly 4 cm is high risk, and a tumor over 4 cm at any site is very high risk.2
  • Depth from 2-6 mm is high risk; depth over 6 mm or invasion beyond subcutaneous fat is very high risk.2
  • Any perineural invasion is high risk. PNI involving a nerve at least 0.1 mm or extending deeper than the dermis is very high risk.2
  • Poor differentiation and lymphovascular invasion are very-high-risk features.2

These examples do not replace the full table. They show why the pathology report and physical examination must be read together.

What do those categories predict?

In a retrospective cohort of 10,196 cSCC tumors used to validate the 2022 NCCN system, very-high-risk tumors had markedly higher adjusted five-year local recurrence, nodal metastasis, distant metastasis, and disease-specific death than low-risk tumors.7

What do those categories predict?
Adjusted 5-year cumulative incidenceLow riskHigh riskVery high risk
Local recurrence0.8%1.5%9.4%
Nodal metastasis0.1%0.5%7.3%
Distant metastasis0.01%0.1%3.9%
Disease-specific death0.1%0.5%10.5%

These are cohort estimates, not personal predictions. They show why very high risk is not decorative wording.

Evidence in numbersEvidence in Numbers: As Risk Rose, Complete Margin Assessment Mattered More

Evidence in Numbers: As Risk Rose, Complete Margin Assessment Mattered More

In the same observational cohort, Mohs or another peripheral and deep en face margin assessment (PDEMA) method was associated with lower adjusted subhazards of local recurrence (SHR 0.65), distant metastasis (SHR 0.38), and disease-specific death (SHR 0.55) than standard wide local excision.7

In plain English, those estimates correspond to 35%, 62%, and 45% lower adjusted subhazards. Because treatment was not randomized, the study shows a strong association rather than proving that the operation alone caused every difference.

How do BCC and cSCC risk features compare?

BCC and cSCC Crosswalk

BCC and cSCC Crosswalk
FactorBCCcSCCWhat it may change
RecurrenceHigh-risk featureHigh-risk featureStronger margin strategy and specialist care
Poor bordersHigh riskHigh riskValue of mapped margin assessment
Aggressive histologyInfiltrative and related patterns raise riskPoor differentiation and aggressive subtypes raise riskMohs or peripheral and deep en face margin assessment (PDEMA), staging discussion, follow-up
DepthConsidered with invasion and anatomyExplicit 2-6 mm and >6 mm/beyond-fat tiersTreatment intensity and prognostic evaluation
Perineural invasionHigh riskHigh or very high risk depending on featuresMargin strategy, imaging/radiation discussion, surveillance
Lymphovascular invasionUncommon; diagnosis-specific interpretationVery-high-risk featureSpecialist and staging discussion
Immune suppressionRaises treatment riskEspecially consequentialLower threshold for specialist care and closer surveillance
Metastatic potentialRareClinically meaningful in high-risk diseasecSCC may require nodal or multidisciplinary planning
Why does melanoma use a different risk rulebook?

Melanoma: Use a Different Rulebook

Melanoma risk is not defined by BCC/cSCC location-size tables.

The core melanoma framework uses:

  • Breslow thickness;
  • microscopic ulceration;
  • lymph-node status;
  • distant metastasis;
  • selected additional pathology and clinical features.10,11

The correct patient action is to move to the Melanoma Treatment Options and melanoma-staging pages. A thin melanoma, melanoma in situ, desmoplastic melanoma, and node-positive melanoma require different decisions. This gateway should not pretend to contain the entire melanoma pathway.

Why does rare-tumor risk begin with the diagnosis?

Rare Tumors: Risk Begins With the Correct Diagnosis

Merkel cell carcinoma, DFSP, sebaceous carcinoma, microcystic adnexal carcinoma, atypical fibroxanthoma, pleomorphic dermal sarcoma, extramammary Paget disease, and sweat-gland cancers cannot be safely placed into a generic BCC/cSCC table.

For a rare tumor, the first risk decision is often diagnostic:

  • Is the pathology correct?
  • Was the sample deep and representative?
  • Does expert dermatopathology review change classification?
  • Does the tumor need imaging, nodal evaluation, systemic therapy, radiation, Mohs, staged excision, or another operation?

Use the rare-tumor branch rather than forcing a familiar cancer's rules onto an unfamiliar disease.

What can a high-risk category change?

What Does High Risk Change?

High-risk features can change several parts of care.

Margin assessment

Current NCCN guidance prefers Mohs or another form of peripheral and deep en face margin assessment for high-risk BCC and for high- or very-high-risk cSCC in appropriate surgical candidates.1,2

That does not make Mohs mandatory for every case. It means the value of a mapped margin rises as the tumor becomes harder to see, more likely to recur, or more dangerous if incompletely removed.

Specialist involvement

A fellowship-trained Mohs surgeon, dermatologic surgeon, dermatopathologist, radiation oncologist, head-and-neck surgeon, surgical oncologist, or medical oncologist may become relevant depending on the diagnosis. The right team follows the problem; it is not the same team for every high-risk cancer.

Staging and nodal evaluation

Selected high-risk cSCC and melanoma cases may require imaging or lymph-node discussion. Those thresholds belong on disease-specific pages.

Reconstruction planning

High-risk tumors in complex anatomy benefit from cancer-clearance planning before major reconstruction. Repair should follow the true defect and the margin strategy.1,2

Follow-up

Higher-risk tumors and immunosuppressed patients generally need closer surveillance. The follow-up page should specify diagnosis-specific schedules rather than using one universal calendar.

Urgency

High risk usually calls for prompt, diagnosis-specific planning, not panic or an assumption that the cancer has spread. It does not create one universal number of days. Rapid growth, recurrence, neurologic symptoms, immune suppression, or high-risk pathology should move specialist review forward.

Features that should accelerate communication with the treating team include:

  • rapid growth;
  • new pain, numbness, tingling, or weakness;
  • fixation to deeper tissue;
  • enlarged nearby lymph nodes;
  • recurrence after prior treatment;
  • immune suppression with a concerning cSCC;
  • pathology showing very-high-risk features.2

This is not an emergency checklist. It is a reason not to let a high-risk result sit in an unowned inbox.

What can a low-risk category change?

What Does Low Risk Change?

Low risk should make treatment more proportionate, not less careful.

Depending on cancer type and site, appropriate options may include:

  • standard excision;
  • ED&C;
  • topical therapy for selected superficial BCC or cSCC in situ;
  • radiation when surgery is not a good fit;
  • observation for selected low-risk BCC in a medically frail patient or a patient with limited life expectancy.1–3,12

A topical medication appropriate for a superficial BCC is not a substitute for Mohs in a recurrent infiltrative nasal BCC. Low risk is the gate that makes simpler treatment possible; it is not permission to treat unlike tumors as interchangeable.

When might treatment or surgery be unnecessary?

When Might Treatment or Surgery Be Unnecessary?

Sometimes a lesion does not require surgery because:

  • the pathology is benign;
  • a selected superficial BCC or cSCC in situ has an evidence-supported nonsurgical option;
  • a selected low-risk BCC is being observed because severe frailty or limited life expectancy makes treatment burden exceed likely benefit;
  • diagnostic uncertainty should be resolved before treatment.1–3,12

This is a physician-guided decision. Wait and see is not appropriate simply because a skin cancer is slow-growing.

Evidence in numbersEvidence in Numbers: Observation Is Selection, Not Neglect

Evidence in Numbers: Observation Is Selection, Not Neglect

In one cohort of 280 BCCs observed in 89 predominantly elderly patients, 124 tumors had at least two suitable measurements. Of those 124 tumors, 58 (46.8%) grew during follow-up. Estimated annual growth was 4.46 mm for infiltrative or micronodular tumors versus 1.06 mm for other subtypes.12 The median patient age was 83, and observation was generally chosen because frailty, major comorbidity, or limited life expectancy changed the balance of benefit and burden.

What should I ask about my skin-cancer risk?

Questions to Ask

  • Which specific features make this tumor low risk, high risk, or very high risk?
  • Are you using a treatment-risk category, a prognostic stage, or both?
  • Does the location or size change the category?
  • Which pathology words change the plan?
  • Was the biopsy deep and representative enough to assess risk?
  • Does immune suppression or prior treatment matter?
  • Would complete mapped margin assessment improve the operation?
  • Do I need disease-specific staging, imaging, nodal evaluation, or closer follow-up?
  • Is a simpler treatment appropriate because the tumor is genuinely low risk?
What do common risk-stratification examples show?

Four Examples

Example 1: Low-risk BCC on the back

A 7-mm primary nodular BCC has clear borders, no aggressive subtype, and no immune-suppression history. Standard excision or another appropriately selected simple treatment may be sufficient. Mohs may add little.

Example 2: High-risk BCC on the nose

A small infiltrative BCC on the nose is high risk despite its diameter. Location, aggressive growth pattern, and limited spare tissue make complete mapped margin assessment especially valuable.

Example 3: Low-risk cSCC on the forearm

A primary, shallow, well-differentiated cSCC under 2 cm with clear borders in an immunocompetent patient may fit standard excision. The treatment page should compare that option with ED&C or Mohs only if the remaining features justify them.

Example 4: Very-high-risk cSCC in a transplant recipient

A recurrent, poorly differentiated cSCC extending beyond fat with PNI is not simply another skin cancer. Mohs/PDEMA may be preferred for the primary margin, but the plan may also require staging, nodal evaluation, radiation, systemic-therapy discussion, and close surveillance.

Frequently asked questions about skin-cancer risk

Frequently Asked Questions

Is low-risk skin cancer still cancer?

Yes. It still requires an appropriate plan. Low risk means a simpler treatment is more likely to be sufficient, not that the diagnosis should be ignored.

Does high risk mean the cancer has spread?

No. A high-risk feature can increase the chance of recurrence or spread without proving that spread has occurred. Stage and risk are different.

Does every high-risk skin cancer need Mohs?

No single rule fits every diagnosis. Current guidance strongly favors Mohs or PDEMA for many high-risk BCCs and high- or very-high-risk cSCCs when surgery is appropriate. Melanoma and rare tumors use diagnosis-specific pathways.1,2

Can a small tumor be high risk?

Yes. A small recurrent tumor, aggressive subtype, poorly defined lesion, cancer on a high-risk site, or tumor with nerve involvement can be high risk despite its diameter.

Can a large tumor still be low risk?

Size is one factor, and current BCC/cSCC tables use specific size thresholds. A large tumor is less likely to remain in the lowest-risk pathway even if other features appear favorable.1,2

Why does immune suppression matter?

The immune system helps control abnormal cells. In transplant recipients and other immunosuppressed patients, cSCC can be more frequent and aggressive. Treatment and surveillance may need to be intensified.

Is BWH the same as NCCN risk?

No. The BWH classification is a cSCC prognostic system. NCCN treatment-risk categories guide treatment selection. They overlap in some features but answer different questions.2,8,9

Where does melanoma fit?

Melanoma follows its own staging system. Breslow thickness, ulceration, lymph-node status, and distant disease are central. Move to the melanoma pathway rather than using BCC/cSCC thresholds.10,11

Where should I go next in the Patient Journey?

Next Steps in the Patient Journey

  • Read the treatment-options page for BCC, cSCC, or melanoma.
  • If the diagnosis is rare or uncertain, use the rare-tumor or pathology second-opinion branch.
  • Continue to Skin Cancer Treatment Options for a general treatment map.
  • Read Do I Need Mohs Surgery? when complete mapped margin assessment may change the plan.
Who reviewed and authored this risk guide?
Portrait of Dr. Thomas L.H. Hocker

About the Author

Dr. Thomas L.H. Hocker is a Harvard- and Mayo Clinic-trained, triple board-certified dermatologist, Mohs surgeon, and dermatopathologist. He is the Founding Director of Dermatologic Surgery at the UMKC School of Medicine and University Health and an Iron Surgeon lecturer at the American Society for Dermatologic Surgery. His work focuses on Mohs surgery for melanoma, complex and rare skin tumors, and aesthetic reconstruction after skin-cancer treatment. He co-authored the best-selling textbook Review of Dermatology and created Skin Trust to give patients and clinicians free access to clear, current, evidence-based education.

Read Dr. Hocker's background and mission

Dr. Hocker earned his bachelor's degree with honors from Yale University, where he was inducted into Phi Beta Kappa. As a Winston Churchill Scholar, he then studied at the University of Cambridge and earned an M.Phil. in Organic Chemistry. He received his M.D. with honors from Harvard Medical School, where his research focused on melanoma genetics. He completed dermatology residency at Mayo Clinic, a dermatopathology fellowship at the University of Michigan, and a Mohs micrographic and reconstructive surgery fellowship at Mayo Clinic. He is board-certified in Dermatology, Dermatopathology, and Mohs Micrographic Surgery.

Dr. Hocker serves as the Founding Director of Dermatologic Surgery at the UMKC School of Medicine and University Health. He is an internationally invited lecturer and speaker who teaches about Mohs surgery for melanoma, complex and rare tumors, dermatopathology, and aesthetic reconstruction after skin-cancer treatment. He has also been selected as an Iron Surgeon lecturer by the American Society for Dermatologic Surgery. He is the co-author of Review of Dermatology, a best-selling dermatology review textbook, and he continues to teach and mentor medical students, residents, and physicians.

Skin Trust exists because Dr. Hocker believes access to excellent medical knowledge should not depend on geography, wealth, or proximity to a major academic center. After training at several of the world's leading institutions, he sees that education as both a gift and a responsibility: to translate current evidence, expert judgment, and hard-won clinical experience into guidance that patients, families, and clinicians can actually use.

The mission is to increase awareness, reduce avoidable suffering, and give every person equal access to trustworthy, up-to-date information that can help them make the best decisions for their life. Skin Trust also extends Dr. Hocker's lifelong commitment to teaching, writing, and mentoring medical students and residents as they build lives and careers of purpose and service.

For Dr. Hocker, this work is also an expression of faith. He regards the opportunities to learn at Yale, Cambridge, Harvard, Mayo Clinic, and the University of Michigan as blessings from God. Teaching, writing, mentoring, and building Skin Trust are ways to pay those blessings forward in service to patients, learners, and the broader community. His faith is the personal motivation to do this work carefully, generously, and with integrity; it is not a condition of using or benefiting from this free resource.

References for this skin-cancer risk guide

References

  1. National Comprehensive Cancer Network. Basal Cell Skin Cancer. Version 2.2026. BCC-2.
  2. National Comprehensive Cancer Network. Squamous Cell Skin Cancer. Version 2.2026. SCC-2.
  3. Kim JYS, Kozlow JH, Mittal B, et al. Guidelines of care for the management of cutaneous squamous cell carcinoma. J Am Acad Dermatol. 2018;78(3):560-578. DOI: 10.1016/j.jaad.2017.10.007. PMID: 29331386.
  4. Kim JYS, Kozlow JH, Mittal B, et al. Guidelines of care for the management of basal cell carcinoma. J Am Acad Dermatol. 2018;78(3):540-559. PMID: 29331385. DOI: 10.1016/j.jaad.2017.10.006.
  5. Rubin AI, Chen EH, Ratner D. Basal-cell carcinoma. N Engl J Med. 2005. PMID: 16306523. DOI: 10.1056/NEJMra044151.
  6. Bisceglia M, Panniello G, Galliani CA, et al. Metastatic basal cell carcinoma of the skin: a comprehensive literature review, including advances in molecular therapeutics. Adv Anat Pathol. 2020;27(5):331-353. PMID: 32618586. DOI: 10.1097/PAP.0000000000000267.
  7. Stevens JS, Murad F, Smile TD, et al. Validation of the 2022 NCCN risk stratification for cutaneous squamous cell carcinoma. JAMA Dermatol. 2023;159(7):728-735. PMID: 37285135. DOI: 10.1001/jamadermatol.2023.1353.
  8. 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. PMID: 24366933. DOI: 10.1200/JCO.2012.48.5326.
  9. 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. PMID: 30969315. DOI: 10.1001/jamadermatol.2019.0032.
  10. Gershenwald JE, Scolyer RA, Hess KR, et al. Melanoma staging: evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(6):472-492. PMID: 29028110. DOI: 10.3322/caac.21409.
  11. National Comprehensive Cancer Network. Melanoma: Cutaneous. 2026.
  12. van Winden MEC, Hetterschijt CRM, Bronkhorst EM, et al. Evaluation of watchful waiting and tumor behavior in patients with basal cell carcinoma: an observational cohort study of 280 basal cell carcinomas in 89 patients. JAMA Dermatol. 2021;157(10):1174-1181. PMID: 34495284. DOI: 10.1001/jamadermatol.2021.3020.
  13. Haws AL, Rojano R, Tahan SR, Phung TL. Accuracy of biopsy sampling for subtyping basal cell carcinoma. J Am Acad Dermatol. 2012;66(1):106-111. PMID: 21798620. DOI: 10.1016/j.jaad.2011.02.042.