Lymphomatoid papulosis: A challenge in the differential diagnosis of ulcerative lesions of the oral mucosa

Abstract

Lymphomatoid papulosis (LyP) is a primary cutaneous CD30+ lymphoproliferative disorder, with a higher incidence in males. Despite its benign and self-resolving clinical course, it exhibits histological features that may mimic other conditions. Mucosal involvement in LyP represents an atypical presentation. We report the case of a 50-year-old man with oral and genital mucosal lesions, initially diagnosed as Behçet's disease. Following a systematic review, 35 additional cases of LyP with mucosal involvement were identified, more frequently in women and with a predilection for the tongue. In 55% of patients, mucosal involvement constituted the initial manifestation of the disease, leading to diagnostic delays and misdiagnosis in one-third of cases. The prognosis was similar to that of classic LyP, with no increased risk of secondary neoplasms. Regarding therapeutic management, methotrexate and topical corticosteroids emerged as the main treatment options.

Resumen

La papulosis linfomatoide (PL) es un trastorno linfoproliferativo cutáneo primario CD30+, con mayor incidencia en varones. A pesar de su curso clínico benigno y autorresolutivo, presenta hallazgos histológicos que pueden simular otros procesos. La afectación mucosa en la PL constituye una forma atípica. Presentamos el caso de un varón de 50 años con lesiones en mucosa oral y genital, inicialmente diagnosticado de enfermedad de Behçet. Tras una revisión sistemática, se identificaron 35 casos adicionales de PL con compromiso mucoso, más frecuente en mujeres y con predilección por la lengua. En el 55% de los pacientes, la afectación mucosa constituyó la primera manifestación de la enfermedad, lo que favoreció retrasos y errores diagnósticos en un tercio de los casos. El pronóstico fue similar al de la PL clásica, sin observarse un mayor riesgo de segundas neoplasias. En el manejo terapéutico, destacaron el uso de metotrexato y corticoides tópicos.

Introduction

Primary cutaneous lymphomas (PCLs) represent the second most common form of extranodal non-Hodgkin lymphoma after gastrointestinal tract lymphomas, with an annual incidence of 1 per 100,000 inhabitants.¹ Among these, mycosis fungoides (MF) and Sézary syndrome (SS) are the most frequent, followed by primary cutaneous CD30+ lymphoproliferative disorders (PC-CD30+ LPDs), which include primary cutaneous anaplastic large cell lymphoma (PC-ALCL) and lymphomatoid papulosis (LyP).¹

LyP, first described in 1968, is a self-limited benign condition, more prevalent in men aged 40–60 years.¹ Clinically, it follows a course characterised by recurrent flares of erythematous–violaceous papules, plaques, or nodules measuring less than 2 cm, which may evolve into vesicular, haemorrhagic, or crusted lesions and subsequently resolve spontaneously. Lesions typically involve the trunk and extremities and less frequently affect acral or facial sites.¹ Mucosal involvement is uncommon (≈3%) but represents a diagnostic challenge, particularly when it constitutes the initial manifestation of the disease. These lesions, involving the oral or genital mucosa, usually present as recurrent ulcerations and pose a complex differential diagnosis with high-grade lymphomas, Epstein–Barr virus (EBV)-positive mucocutaneous ulcer, and other inflammatory conditions.

We report the case of a patient with LyP presenting with mucosal involvement at disease onset, together with a systematic review of the literature aimed at analysing the clinical, diagnostic, and therapeutic characteristics, as well as the main differential diagnoses, of this atypical form of LyP.

Case report

A 50-year-old man initially presented to a private clinic with cutaneous and oral and genital mucosal lesions. He was diagnosed with Behçet's disease (HLA-B51 positive) and treated with azathioprine and colchicine, with poor response. The cutaneous lesions were recurrent and self-regressing. Owing to an unfavourable clinical course, a skin biopsy was performed, revealing findings suggestive of a lymphoproliferative disorder, prompting referral to our centre in February 2020.

On examination, the patient presented with erythematous–violaceous papules on the penis and scrotum, a perioral pustular lesion, and an ulcer involving the perianal mucosa. In addition, he provided images of previous flares showing lesions in the oral cavity (Fig. 1). Biopsies were obtained from the penile and perioral lesions. Both specimens demonstrated a mixed inflammatory infiltrate in the subepithelial tissue, with a perivascular and interstitial distribution, composed of large, atypical cells admixed with small lymphocytes and eosinophils. In the perioral biopsy, the lesion was ulcerated and showed a denser and more diffuse lymphoid proliferation. On immunohistochemical analysis, the large, atypical cells were positive for CD3, CD5, CD30, CD4, and MUM1, and negative for CD79a, CD20, CD7, CD8, ALK1, and EBER (Fig. 1). Based on these findings, a diagnosis of primary cutaneous CD30-positive lymphoproliferative disorder, lymphomatoid papulosis type, was established. The patient was started on methotrexate and topical corticosteroids, with good tolerance. He experienced alternating periods of remission and disease flares, during which methotrexate dose was temporarily adjusted. At the most recent follow-up visit (January 2025), a suspicious lesion consistent with possible LyP was noted on the lower abdomen. After four years of follow-up, the patient has not developed lymphadenopathy, tumoural lesions, or patch- or plaque-type lesions suggestive of MF.

Fig. 1.

(A) Clinical photograph. An 18 mm oral ulcer on the left buccal mucosa observed during one of the disease flares. (B) At low magnification (H&E, ×1), a punch biopsy of the oral mucosa shows ulceration and expansion of the lamina propria by a cellular proliferation. (C) At higher magnification (H&E, ×20), this proliferation is heterogeneous, comprising scattered large cells with anaplastic features interspersed among small, mature-appearing lymphocytes, histiocytes, and eosinophils. (D) Immunohistochemical analysis demonstrates CD30 expression in the anaplastic cell population (×20). This population is also positive for CD3 (E, ×1.5) and CD4 (G, ×1.5), and negative for CD7 (F, ×1.5) and CD8 (H, ×1.5).

Discussion

LyP, together with PC-ALCL, constitutes the group of primary cutaneous CD30-positive lymphoproliferative disorders, which represent the second most frequent group of T-cell lymphoproliferative disorders after MF/SS. There is growing interest in this condition, not only because of its rarity but also due to the diagnostic challenges it poses in daily clinical practice. Although it follows an indolent course, LyP displays histological features that overlap with those of other aggressive lymphomas and/or inflammatory or infectious processes.¹ It typically affects middle-aged men, although paediatric cases have also been reported.¹ Clinically, it is characterised by recurrent flares of papulonodular lesions with a tendency towards spontaneous resolution, predominantly involving the trunk and extremities. Chronic evolution, lesion polymorphism, and the absence of systemic symptoms are characteristic features of the disease.

Mucosal involvement in LyP, as observed in our case, is exceptional and represents a significant diagnostic challenge in clinical practice for both dermatologists and dermatopathologists. To date, including the present case, 36 adult cases of LyP with mucosal involvement have been identified, comprising 15 men and 21 women, and are summarised in Table 1. Interestingly, this distribution shows a higher frequency in women, in contrast to the male predominance typically observed in classic LyP. In the reported cases, mucosal involvement was most often multifocal, with a predilection for the tongue (61%, n = 22)^2–17 and the oral mucosa (39%, n = 14).^{2,4,13–15,17–20} Less frequently, lesions have been reported on the labial mucosa (25%, n = 11),^{2,13,14,17,21,22} genital mucosa (8%, n = 3),^13,21 and palatal mucosa (2%, n = 1).²³ Despite the striking nature of mucosal involvement due to its location and ulcerated morphology, no negative impact on prognosis has been demonstrated. LyP retains an indolent course, with a reported five-year overall survival of 100%.¹

Table 1.

Summary of documented cases in the literature to date of LyP with mucosal involvement.

Author(s)	Year	Age	Sex	Location of the lesion	Characteristics of the lesion	Cutaneous involvement	Treatment and outcome
Thomsen et al.^20	1972	83	F	Hands and oral mucosa	Blistering lesions of the oral mucosa	Yes	SR, unsuccessful treatment
Sioutos et al.^3	1997	60	M	Tongue	Tumour at the base of the tongue	No	CT and RT. Relapses on trunk and limbs treated with PUVA
Kato et al.^10	1998	34	M	Tongue and skin	Reddish depressed nodules on the tongue and papulovesicles on the extremities	Yes	Interferon (INF)-alpha2a prevented recurrences
Sciubba et al.^12	2000	60	F	Tongue	Multiple ulcers	Yes (previous)	SR
Chimenti et al.^21	2001	38	F	Oral commissure (lip) and vulva	Nodular and papulovesicular lesions	Yes (previous)	SR
Pujol et al.^11	2005	5233	FM	Tongue, oral commissureTongue	Ulcerated nodule and papule	Yes (previous)	Initially MTX and spontaneous remission of mucosal lesion/Topical corticosteroids and MTX
Serra-Guillén et al.^7	2007	67	M	Tongue	Ulcers	Yes (previous)	Oral and topical corticosteroids and MTX
Agarwal et al.^4	2008	4636	F	Tongue and oral mucosa	Ulcers	No	SR
Allabert et al.^13	2008	38713766	FFFM	Lip and nasal mucosaTongue and lipVulva, tongue and lipOral mucosa	Ulcers	2 yes (previous),2 no	1 treated with MTX without relapses; 2 with corticosteroids and 1 untreated
De-Misa et al.^16	2009	72	M	Tongue	Red papule with central yellow ulcer and raised borders	No	SR
Saggini et al.^5	2010	19	F	Tongue	Ulcerated nodule	Yes (previous)	Untreated
Booken et al.^15	2013	85	M	Tongue, oral vestibule	Ulcerated nodules with raised border	No, later on fingers	Untreated
Kempf et al.^17	2013	746266	MMF	Oral cavity, lipLip, noseLip, oral mucosa, tongue, cheek	Ulcers	NoYesYes	Topical corticosteroidsSurgeryUntreated
Benslama et al.^18	2015	3263	F	Oral mucosa	Ulcer	No	SR
Schwartz et al.^14	2017	4271345462	MFFFF	Tongue (4), oral mucosa (2), lip (2)	Ulcers	No	3 untreated, 1 with corticosteroids, 1 with CT
Jessri et al.^19	2017	63	F	Oral mucosa	Ulcer	Yes (previous)	Corticosteroids
Bretstzajn et al.^6	2019	39	F	Tongue	Ulcers	No	SR. Recurrences with MTX on palate, finger, eyelids, nasal cavities, scalp
Machan et al.^2	2020	2155	MM	Fingers, oralNose, tongue, lip	Ulcerated nodules with necrotic centre/Papules	Yes (previous)	Topical corticoisteroids and MTX. Scrotal relapse. SR followed by recurrences
Narahara et al.^22	2021	25	F	Right upper lip	Slightly elastic exophytic mass	No	Topical corticosteroids
Silveira et al.^23	2023	24	M	Hard palate in molar region	Ulcer	No	Remission with topical corticosteroid without relapses
Barbeiro et al.^9	2024	60	M	Tongue	Ulcer	No	Remission with topical corticosteroid without relapses
Gallo et al.^8	2024	50	F	Tongue	Ulcers	No	SR
Our case	2025	50	M	Oral and genital mucosa	Ulcers and erythematous–violaceous papules	Concurrent cutaneous involvement: beard area, perinasal, perioral, perianal	Recurrent lesions treated with MTX and topical corticosteroids

SR: spontaneous resolution, CT: chemotherapy, RT: radiation therapy, MTX: methotrexate.

The timing of mucosal involvement during the course of the disease is heterogeneous, as it may occur either at disease onset or during follow-up. Diagnosis is particularly challenging when mucosal involvement represents the initial clinical manifestation. In this review, more than half of the patients (55%, n = 20) presented with exclusively mucosal involvement at disease onset, without associated cutaneous lesions.^{3,4,6,8,9,13–18,22,23} In contrast, simultaneous involvement of the skin and mucosa at the time of diagnosis was observed in only 4 of the 36 cases, including the present case, which facilitated clinicopathological correlation in most of them. These cases included a woman with lesions on the hands and oral mucosa, a man with cutaneous and lingual involvement, another patient with labial and nasal involvement, and a woman with concurrent lesions of the lip, oral mucosa, tongue, and cheek.^10,17,20 Finally, in 11 patients with a previous diagnosis of LyP, mucosal involvement developed later during follow-up.^{2,5,7,11–13,19,21}

As in the classic cutaneous presentation, cases of LyP with mucosal involvement give rise to a broad spectrum of differential diagnoses. These manifestations may mimic inflammatory or infectious processes, as well as aggressive malignant neoplasms, thereby entailing a considerable risk of diagnostic error. Such errors may have significant clinical consequences, particularly when they lead to the initiation of inappropriate or unnecessary treatments. Indeed, up to one third of the reviewed patients with mucosal onset (n = 7)^3,6,14,22 were initially diagnosed with conditions other than LyP. Among the most frequent diagnostic errors were other non-Hodgkin lymphomas.³ One illustrative example is a patient initially diagnosed with PC-ALCL and treated with chemotherapy and radiotherapy, with the consequent risk of short- and long-term toxicity.³ In another case, a patient diagnosed with high-grade non-Hodgkin lymphoma and treated with chemotherapy was subsequently reclassified as having LyP following histological examination of the mucosal lesions.²¹ Additionally, four patients with exclusively mucosal onset were misdiagnosed with peripheral T-cell lymphoma.¹⁴ Of particular note is the case of a 54-year-old woman who, after two years, developed additional cutaneous lesions suggestive of MF. Re-evaluation of the lingual biopsy allowed the correct diagnosis of LyP to be established. Eleven years after disease onset, the patient developed new lesions consistent with MF.¹⁴ The association between LyP and MF has been described in patients with classic LyP without mucosal involvement and does not appear to influence prognosis.²⁴ Furthermore, molecular studies have demonstrated the presence of an identical clone in MF and LyP lesions, suggesting a common origin.²⁵

To a lesser extent, some cases of LyP with mucosal involvement were initially interpreted as infectious or reactive conditions. One example is a woman with mucosal aphthae that were diagnosed as recurrent herpetic lesions.⁶ In another case, the patient was evaluated by a different department and received two consecutive incorrect diagnoses, the latter being pyogenic granuloma.²² Notably, no cases of LyP with mucosal involvement initially diagnosed as Behçet's disease have been identified in the literature. Behçet's disease may present with recurrent oral and genital ulcerations, which can be misleading in this context. However, HLA-B51 positivity alone is not diagnostic of Behçet's disease, as it can also be found in healthy individuals, in whom it acts as a genetic marker of increased susceptibility to disease development but does not imply diagnosis or certainty of future disease.²⁶ This is exemplified by our case, in which the absence of systemic involvement, the self-limited course, the lack of response to immunosuppressive therapy, and the absence of other diagnostic criteria support exclusion of this condition. Conversely, it has been reported that the HLA-B51 haplotype may confer increased susceptibility to lymphoproliferative disorders, including non-Hodgkin lymphomas, although the absolute risk is low. However, there is currently no evidence in the medical literature specifically linking HLA-B51 to susceptibility to LyP.²⁷

As previously mentioned, LyP follows a chronic and relapsing course. Several management options are available, ranging from watchful waiting in mild cases to the use of methotrexate, topical corticosteroids, or other immunomodulatory therapies in more extensive or symptomatic disease.¹ In cases with mucosal involvement, a similar heterogeneity in clinical management is observed, with a stepwise, conservative approach and individualised monitoring. Approximately 53% of patients with mucosal LyP experienced spontaneous remission in the absence of treatment, consistent with the characteristic self-resolving nature of this condition. Nevertheless, other cases have been reported in which patients were treated with oral or topical corticosteroids and methotrexate, either immediately after diagnosis or during relapses.^{2,7,9,11,13,14,17,19,22,23} Additionally, one case treated with interferon alpha has been described,¹⁰ as well as another managed surgically.¹⁷ In our patient, following the diagnosis of LyP, treatment with methotrexate and topical corticosteroids was initiated, with a good clinical response.

It is noteworthy that, although between 10% and 20% of patients with LyP develop a second haematological malignancy, no association with another lymphoma has been demonstrated in our patient after four years of follow-up. In the review, 94% of patients with mucosal involvement likewise did not develop a second lymphoma; only two cases were reported with features consistent with MF. One of these has been discussed previously.¹⁴ In the other, a glossectomy was performed, and subsequent biopsies showed a pattern resembling that of a large cell lymphoma.²⁰ Therefore, despite the generally benign course of this condition, the importance of continued long-term follow-up should be emphasised, allowing early detection should other lymphomas arise, some of which may have a more aggressive clinical course.

In summary, LyP with mucosal involvement represents a diagnostic challenge, with clinicopathological correlation and a multidisciplinary approach being essential, particularly in atypical presentations such as the case described. Although the clinical course is usually favourable, prolonged follow-up is required owing to the low, yet present, risk of association with other neoplasms. Important gaps remain in our understanding of its epidemiology, pathogenesis, and management, as most available data derive from case reports and small case series. Recognition of this presentation is crucial to guide the diagnostic process appropriately from the outset and to avoid unnecessary treatments.

Ethical considerations

Not applicable.

Informed consent

Written informed consent was obtained.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of interest

The drafting of this paper was supported by the Horacio Oliva Grant.