Atypical fibroxanthoma (AFX) and pleomorphic dermal sarcoma (PDS) represent a spectrum of UV-induced cutaneous sarcomas arising on sun-damaged skin of elderly patients. AFX is confined to the dermis with excellent prognosis (<1% metastasis), while PDS extends into the subcutis and/or harbors adverse histologic features (necrosis, lymphovascular invasion, perineural invasion) carrying 10-20% metastatic potential. Both are diagnoses of exclusion. The hallmark IHC finding is negativity for all lineage-specific markers (cytokeratins, S-100, desmin, Melan-A), requiring systematic exclusion of spindle cell SCC, melanoma, leiomyosarcoma, and angiosarcoma. Mohs micrographic surgery is the treatment of choice for both AFX and PDS, offering superior margin control and local recurrence rates. For PDS with high-risk features, adjuvant radiation therapy and sentinel lymph node biopsy should be considered. This article provides a thorough overview of the AFX-PDS spectrum with emphasis on histopathologic diagnosis, IHC workup, Mohs surgical technique, and management algorithms.
By Dr. Yehonatan Kaplan (M.D., Fellow ACMS)·Published: 2025-03-01·Updated: 2026-03-15·Reviewed: 2026-03-07
Atypical fibroxanthoma (AFX) and pleomorphic dermal sarcoma (PDS) are UV-induced cutaneous sarcomas that are now recognized as representing a single biologic spectrum rather than distinct entities. The WHO Classification of Skin Tumours (5th edition, 2023) classifies both under the category of undifferentiated/unclassified sarcomas of the skin. AFX represents the superficial, dermis-confined, favorable end of the spectrum, while PDS represents the deeper, more aggressive end with extension into the subcutaneous fat and/or the presence of adverse histologic features. The nomenclature has evolved significantly: historically AFX was considered a benign or low-grade tumor with universally excellent prognosis, but the recognition that tumors with identical cytomorphology but deeper invasion or adverse features carry significant metastatic potential led to the distinction of PDS as a separate prognostic category. The estimated incidence of AFX/PDS combined is approximately 0.3-0.5 per 100,000 per year, though this is likely underestimated due to historical misdiagnosis and inconsistent nomenclature. Both AFX and PDS arise predominantly on chronically sun-damaged skin of the head and neck in elderly men. The typical patient is a male over 70 years old with extensive actinic damage. The scalp, ears, nose, and cheeks are the most common sites, with over 80% of cases occurring on the head and neck. Unlike DFSP, which affects younger adults on the trunk, the AFX-PDS spectrum is strongly associated with cumulative UV exposure and occurs almost exclusively in the elderly. Immunosuppression, prior radiation therapy, and xeroderma pigmentosum are additional risk factors.
Clinical Presentation
AFX and PDS typically present as a rapidly growing, dome-shaped, pink to red nodule on sun-damaged skin of the head and neck. The lesion often develops over weeks to a few months, distinguishing it from the more indolent growth pattern of BCC. Ulceration is common and may be the presenting complaint, as the rapidly growing tumor outstrips its blood supply. The size at presentation is typically 1-2 cm for AFX, though PDS may present at larger sizes. The surface may be smooth, exophytic, polypoid, or crusted with ulceration. Bleeding is a common symptom, often prompting clinical attention. The clinical appearance is nonspecific and closely mimics other common tumors on sun-damaged skin. The clinical differential diagnosis includes squamous cell carcinoma (the most common clinical misdiagnosis), basal cell carcinoma (nodular type), keratoacanthoma, amelanotic melanoma, Merkel cell carcinoma, and cutaneous metastasis. There are no pathognomonic clinical features that distinguish AFX/PDS from these entities, making biopsy mandatory for any rapidly growing nodule on actinically damaged skin of elderly patients. On examination, the surrounding skin typically shows extensive photodamage with solar elastosis, actinic keratoses, and often a history of prior skin cancers. This field cancerization context is important diagnostically. Regional lymphadenopathy is uncommon at presentation for AFX but should be assessed for PDS, particularly in tumors with high-risk features.
Histopathology & Immunohistochemistry
AFX and PDS share identical cytomorphologic features. Both are characterized by a dermal proliferation of highly pleomorphic spindle cells, epithelioid cells, and bizarre multinucleated giant cells arranged in a haphazard or storiform pattern. The mitotic rate is typically high (>10 mitoses per 10 HPF), and atypical mitotic figures are frequent. The degree of nuclear pleomorphism is striking, often more dramatic than in the tumors being excluded (SCC, melanoma, leiomyosarcoma), and paradoxically this extreme pleomorphism can be a diagnostic clue favoring AFX/PDS. The overlying epidermis frequently shows epidermal collarette formation at the periphery of the tumor, and the adjacent dermis shows pronounced solar elastosis. The tumor stroma may contain inflammatory infiltrates and areas of hemorrhage. The most critical diagnostic principle for AFX/PDS is that it is a diagnosis of EXCLUSION. There is no single positive immunohistochemical marker that confirms the diagnosis. Instead, the diagnosis requires systematic exclusion of all specific lineage-differentiated tumors that can mimic this histologic pattern. The hallmark IHC finding is negativity for all lineage-specific markers.
Mandatory IHC Panel for Diagnosis
A thorough IHC panel must be performed to exclude mimics before a diagnosis of AFX or PDS can be rendered. The essential panel includes: pan-cytokeratins (AE1/AE3, CK5/6, and ideally p40 or p63). These MUST be negative to exclude spindle cell (sarcomatoid) squamous cell carcinoma, which is the most important and most commonly missed mimic. S-100 protein and SOX10. Both must be negative to exclude desmoplastic or spindle cell melanoma. Melan-A/MART-1. Negative, to further exclude melanoma. Desmin and smooth muscle actin (SMA). Desmin must be negative to exclude leiomyosarcoma; SMA may show focal weak positivity in AFX/PDS (myofibroblastic differentiation) but strong diffuse SMA positivity should raise concern for leiomyosarcoma. CD31 and ERG. Both must be negative to exclude angiosarcoma. CD34. Typically negative in AFX/PDS (helps exclude DFSP and angiosarcoma). CD10 is frequently positive in AFX/PDS and historically was considered a supportive marker, but CD10 positivity is non-specific and occurs in many other tumors. Procollagen-1 is another supportive marker of fibroblastic differentiation but is also non-specific. P53 is frequently overexpressed in AFX/PDS, reflecting UV-driven TP53 mutations, but this is not diagnostically specific.
IHC Marker
AFX/PDS
Spindle SCC
Desmoplastic Melanoma
Leiomyosarcoma
Angiosarcoma
CK (AE1/AE3)
NEGATIVE
POSITIVE
Negative
Negative
Negative
p40 / p63
Negative
POSITIVE
Negative
Negative
Negative
S-100
Negative
Negative
POSITIVE
Negative
Negative
SOX10
Negative
Negative
POSITIVE
Negative
Negative
Melan-A
Negative
Negative
Variable
Negative
Negative
Desmin
Negative
Negative
Negative
POSITIVE
Negative
SMA
Neg/focal weak
Negative
Negative
POSITIVE (diffuse)
Negative
CD31 / ERG
Negative
Negative
Negative
Negative
POSITIVE
CD10
Positive (non-specific)
Variable
Variable
Variable
Variable
Procollagen-1
Positive (non-specific)
Negative
Negative
Negative
Negative
Distinguishing AFX from PDS on Histopathology
Once the diagnosis of an undifferentiated pleomorphic sarcoma of the skin is established by IHC exclusion, the next critical step is determining whether the tumor meets criteria for AFX (favorable) or PDS (unfavorable). This distinction is made entirely on histopathologic grounds and has profound prognostic and therapeutic implications. AFX is defined as a tumor confined to the dermis. The deep margin does not extend into the subcutaneous fat. PDS is defined by one or more of the following adverse features: invasion into the subcutaneous fat (beyond the dermis), tumor necrosis, lymphovascular invasion (LVI), perineural invasion (PNI), or large tumor size (typically >2 cm). Some authorities also include deep invasion into fascia, muscle, or bone as criteria for PDS. The presence of ANY single adverse feature is sufficient to reclassify what would otherwise be an AFX as a PDS. This underscores the importance of adequate biopsy depth. A superficial shave biopsy that does not sample the deep margin cannot reliably distinguish AFX from PDS, because subcutaneous invasion may be present below the biopsy plane.
AFX vs. PDS: Distinguishing Features & Prognosis
The distinction between AFX and PDS is the most clinically consequential classification decision in this tumor spectrum, as it determines prognosis, follow-up intensity, and the need for adjuvant therapy. AFX, by definition confined to the dermis and lacking adverse histologic features, behaves as a low-grade tumor with an excellent prognosis. The local recurrence rate after complete excision is approximately 5-10%, and the metastatic rate is less than 1%. In contrast, PDS is a significantly more aggressive tumor with a local recurrence rate of 10-20% after standard excision, a metastatic rate of 10-20%, and disease-specific mortality reported at 5-10% in published series. The most common sites of distant metastasis for PDS are the lungs and regional lymph nodes, though metastases to bone, brain, and other visceral sites have been reported. The dramatic difference in biologic behavior between AFX and PDS underscores the importance of accurate histopathologic classification and the danger of lumping both entities together under the AFX label.
Feature
AFX (Favorable)
PDS (Unfavorable)
Depth of invasion
Confined to dermis
Extends into subcutis or deeper
Tumor necrosis
Absent
Present
Lymphovascular invasion
Absent
Present
Perineural invasion
Absent
Present
Typical size at presentation
<2 cm
Often >2 cm
Local recurrence rate
5-10%
10-20%
Metastatic potential
<1%
10-20%
Disease-specific mortality
Negligible
5-10%
Most common metastatic sites
N/A (extremely rare)
Lung, regional lymph nodes
Recommended surgical approach
Mohs. Curative in vast majority
Mohs with wider initial margins; consider adjuvant RT
SLNB consideration
Not indicated
Consider for high-risk features
Adjuvant RT
Not indicated
Consider for adverse features
Follow-up intensity
Standard skin cancer follow-up
Close surveillance with imaging consideration
Mohs Micrographic Surgery
Mohs micrographic surgery (MMS) is the treatment of choice for both AFX and PDS. The rationale for Mohs in AFX/PDS is particularly compelling: these tumors arise on chronically sun-damaged skin of the head and neck. The same field cancerization environment that produces BCCs and SCCs. Making subclinical extension along sun-damaged tissue planes a genuine concern. Mohs provides complete peripheral and deep margin assessment (100% margin evaluation) in a setting where clinical margin estimation is unreliable due to the field cancerization and actinic damage of surrounding skin. Published local recurrence rates after Mohs for AFX are approximately 0-2%, compared to 5-10% after standard excision. For PDS, Mohs achieves local recurrence rates of approximately 2-8%, compared to 10-20% after wide local excision. These superior recurrence rates reflect the advantage of complete margin assessment in a tumor that can have irregular subclinical extension. Additionally, Mohs offers critical tissue conservation on the head and neck, where the majority of AFX/PDS tumors occur. The scalp, nose, ears, and periauricular areas are common sites where standard excision with 1-2 cm margins may produce large defects requiring complex reconstruction.
Mohs Technique Considerations for AFX/PDS
Mohs for AFX/PDS requires several important technical considerations that differ from routine Mohs for BCC/SCC. First, frozen section interpretation of the highly pleomorphic spindle cells of AFX/PDS can be challenging. The cells are large and atypical but do not form the discrete nests or islands seen in BCC/SCC, making them harder to detect at the margin on frozen sections. The bizarre nuclear pleomorphism can also raise concern for artifact on frozen sections. Second, it is essential to confirm the diagnosis on the Mohs layers themselves with an IHC panel. Because AFX/PDS is a diagnosis of exclusion, performing a limited IHC panel on the Mohs frozen or rush permanent sections (at minimum: pan-cytokeratin AE1/AE3 and S-100) helps ensure that a spindle cell SCC or desmoplastic melanoma is not being treated under the AFX/PDS label. Some Mohs surgeons perform IHC on the first Mohs layer and/or on the central debulking specimen. Third, for PDS specifically, the central debulking specimen should be submitted for permanent vertical sections. Similar to the approach used for MCC. To provide accurate assessment of depth, LVI, PNI, and necrosis for definitive classification and adjuvant therapy planning.
Margin Approach: AFX vs. PDS
For AFX (dermis-confined, no adverse features), standard Mohs technique with sequential layer excision until clear margins are achieved is sufficient. Narrow initial margins (2-3 mm) around the clinically apparent tumor are typically adequate because AFX tends to be relatively well circumscribed. For PDS, a more aggressive margin approach is warranted given the higher recurrence and metastatic risk. Consider starting with wider initial margins (5-10 mm) for the first Mohs layer, and excise the deep margin to at least the subcutaneous fat-fascia junction. Some authors advocate for a modified "slow Mohs" approach for PDS using rush permanent sections with IHC rather than frozen sections alone, to improve the accuracy of margin assessment for this challenging histologic pattern.
Treatment Algorithm & Adjuvant Therapy
The treatment algorithm for AFX/PDS depends critically on the distinction between the two entities and the presence or absence of adverse histologic features. For AFX (dermis-confined, no adverse features), Mohs micrographic surgery alone is curative in the vast majority of cases. No adjuvant therapy is required, and standard post-treatment surveillance is sufficient. For PDS, the management approach is more individualized and should incorporate multidisciplinary discussion.
Management Decision
AFX
PDS
Mohs micrographic surgery
Treatment of choice. Curative
Treatment of choice. Wider initial margins
Standard excision (if Mohs unavailable)
1-2 cm margins
2-3 cm margins to fascia
Adjuvant radiation therapy
Not indicated
Consider for multiple adverse features, close/positive margins
Sentinel lymph node biopsy
Not indicated
Consider for LVI, large tumors, multiple adverse features
For confirmed AFX (dermis-confined, no necrosis, no LVI, no PNI): Mohs micrographic surgery is the treatment of choice. If Mohs is unavailable, wide local excision with 1-2 cm margins is acceptable. Complete excision with clear margins is curative in >95% of cases. No adjuvant radiation therapy is indicated. No sentinel lymph node biopsy is indicated. No imaging or staging workup is required. Follow-up consists of routine skin cancer surveillance every 6-12 months, with attention to the surgical site and full skin examination for new primary tumors given the field cancerization risk.
PDS Treatment Algorithm
For PDS (subcutaneous invasion and/or adverse features): Mohs micrographic surgery with wider initial margins (5-10 mm) is recommended. Consider submitting the debulking specimen for permanent sections to guide adjuvant therapy. Adjuvant radiation therapy (RT) should be considered for tumors with multiple adverse features (necrosis + LVI, PNI, large size >2 cm), positive or close margins after excision, or recurrent disease. RT doses of 50-66 Gy delivered over 5-7 weeks are typically used, similar to protocols for high-risk SCC. Sentinel lymph node biopsy (SLNB) is increasingly recommended for PDS with adverse features, particularly LVI and large tumor size, though no formal guidelines exist and the evidence base is limited to case series. For PDS with LVI or tumors >2 cm, baseline imaging (CT chest and regional nodal basin) should be considered for staging. There is no established systemic chemotherapy regimen for metastatic PDS. Case reports and small series have described responses to immune checkpoint inhibitors (pembrolizumab, nivolumab), which is biologically plausible given the high mutational burden driven by UV damage. For metastatic PDS, referral to medical oncology for consideration of immunotherapy or enrollment in clinical trials is recommended.
Immunotherapy for Metastatic PDS
The high tumor mutational burden (TMB) of AFX/PDS, driven by extensive UV-induced DNA damage, provides a strong biologic rationale for immune checkpoint inhibitor therapy in metastatic PDS. Multiple case reports have demonstrated clinical responses to pembrolizumab (anti-PD-1) in metastatic PDS, with some patients achieving durable partial or complete responses. This is analogous to the responsiveness of UV-associated Merkel cell carcinoma and advanced cutaneous SCC to checkpoint inhibitors. However, the evidence is limited to case reports and small case series. No prospective clinical trials have been completed. For patients with unresectable or metastatic PDS, multidisciplinary tumor board discussion is recommended. Referral to medical oncology for consideration of checkpoint inhibitor therapy (pembrolizumab or nivolumab) is appropriate. Traditional cytotoxic chemotherapy has shown limited efficacy for metastatic undifferentiated pleomorphic sarcomas in general, and no specific regimen is established for PDS.
Prognosis & Follow-Up
The prognosis of AFX is excellent with appropriate surgical management. Local recurrence rates after Mohs are 0-2%, and metastasis occurs in fewer than 1% of cases. Disease-specific mortality from true, dermis-confined AFX is negligible. Patients with AFX can be reassured about the favorable prognosis while being counseled about the importance of ongoing skin cancer surveillance, given that the same UV-damaged skin is at risk for developing additional primary skin cancers (BCC, SCC, new AFX/PDS, or melanoma). The prognosis of PDS is significantly less favorable. Local recurrence rates after excision range from 10-20% (lower with Mohs, approximately 2-8%). Metastasis occurs in 10-20% of PDS cases, with the lungs and regional lymph nodes being the most common metastatic sites. Disease-specific mortality for PDS is reported at 5-10% in published series. Adverse prognostic factors within PDS include tumor necrosis, lymphovascular invasion, perineural invasion, tumor size >2 cm, depth of invasion beyond the subcutis (into fascia or muscle), and positive surgical margins. Recurrence of PDS, both local and distant, typically occurs within the first 2 years after treatment, though late recurrences have been reported.
Outcome
AFX
PDS
Local recurrence (Mohs)
0-2%
2-8%
Local recurrence (WLE)
5-10%
10-20%
Metastatic rate
<1%
10-20%
Disease-specific mortality
Negligible
5-10%
Median time to recurrence
N/A (rare)
Within first 2 years
Follow-up frequency
q6-12 months
q3-6 months x 2 years, then q6-12 months
Imaging surveillance
Not required
Consider annual CT chest x 2-3 years
Follow-Up Recommendations
For AFX: Clinical examination of the surgical site and full skin examination every 6-12 months. No routine imaging is required. Patient education about self-examination of the surgical scar for new nodularity or firmness. Emphasis on sun protection and skin cancer screening, as these patients are at risk for additional primary skin cancers due to field cancerization. For PDS: More intensive surveillance is warranted given the metastatic risk. Clinical examination every 3-6 months for the first 2 years, then every 6-12 months. Regional lymph node examination at each visit. Consider baseline and periodic chest imaging (CT chest annually for 2-3 years) for PDS with LVI, large tumor size, or other high-risk features. Patient education about signs of local recurrence and regional lymphadenopathy. For patients with metastatic PDS, surveillance is dictated by the medical oncologist based on sites of involvement and treatment response.
The Critical Pitfall: AFX Misdiagnosis
The single most important clinical pitfall in managing the AFX-PDS spectrum is misdiagnosing PDS as AFX. This occurs when the initial biopsy is superficial and does not sample the deep margin, when the pathologist does not evaluate or report on adverse features (necrosis, LVI, PNI), or when the debulking or excision specimen is not reviewed for features that would reclassify the tumor. Historical series reporting metastasis rates of 5-10% for "AFX" almost certainly included cases that would be classified as PDS under current criteria. To avoid this pitfall: always ensure an adequate depth biopsy (punch or incisional extending to subcutis), require the pathology report to explicitly address depth of invasion and the presence or absence of necrosis, LVI, and PNI, and review the definitive excision or debulking specimen for features that would reclassify an AFX as PDS. When a tumor initially diagnosed as AFX recurs, always re-evaluate the original specimen and the recurrence specimen for PDS features.