Patients and methods

PATIENTS

Between 1983 and 1997, 67 of 155 new LCH patients seen in our haematology/oncology department had skin involvement. In 20 of them the degree and type of skin involvement were considered to justify topical mustine treatment9 because they were sufficiently severe (pruritic, ulcerating, or fistulating) and unresponsive to topical and oral steroids. Medical records of these 20 patients, who were followed up to 1997–99, were reviewed in detail. LCH was diagnosed according to standard histological criteria24 25 at a mean age of 18.2 months (range 2–108, median 12.5). Using criteria proposed by the coordinators of the LCH-1 Treatment Protocol,24 the average extent of skin involvement was 16.4% of the body surface (range 3–50%, median 15%). Lesions were most common in the scalp and inguinal region (in 90% and 85% of patients, respectively). Fifteen patients also had bone involvement, diabetes insipidus (DI) was diagnosed in 10, and various other organs (lymph nodes, lung, bone marrow, CNS, gastrointestinal tract, gingiva, and conjunctiva) were involved in a total of 14 children. Fifteen patients had also been treated with prednisolone or a variety of cytostatic drugs (table 1).

DETAILS OF MUSTINE TREATMENT

Nitrogen mustard powder (Boots Co., Nottingham, UK) was diluted with water to a final concentration of 200 mg/l and selectively applied to the involved skin with appropriately sized watercolour paintbrushes, as described previously.9 The initial average dose was 2–3 mg nitrogen mustard (mustine) per day. After 10 minutes, the mustine was rinsed off. The solution was applied once daily until the skin lesions started to regress. The frequency of application was then reduced to every other day, then every third day, then once weekly until complete clearance. Treatment was begun in hospital and, after appropriate training, continued by the child's parents at home. The total amount of mustine used was calculated from the number of units dispensed by our hospital pharmacy. The total dose per square metre of body surface was calculated from the amount of drug versus body weight and length, respectively, at the start of mustine treatment (m² body surface = √(body weight × body length) ÷ 3600).

FOLLOW UP INVESTIGATIONS

Clinical follow up examinations were performed by either of two physicians (PHH, VRN) at an average of 8.3 years (range 4–174 months, median 114 months) after completion of topical mustine treatment. The outcome was classified into “active” and “non-active” disease according to the disease state definition of the Histiocyte Society.24 LCH is regarded as non-active when all signs and/or symptoms have resolved. Active disease is subclassified into “regressing disease” (regression of signs and/or symptoms without appearance of new lesions), “stable disease” (persistence of signs and/or symptoms of disease without appearance of new lesions), and “progressive disease” (progression of signs and/or symptoms of disease initially detected and/or reappearance of old and/or appearance of new lesions). Some patients with non-active disease had residual handicaps from previously active LCH, for example, diabetes insipidus, growth hormone deficiency, or deafness.

Results

DETAILS OF MUSTINE TREATMENT

The 20 children received a total of 34 courses of topical mustine. Average time to clearance (at least 90% reduction in size and severity of skin lesions) was 21 days (range 7–42; table 1). After at least 90% clearance was obtained, treatment for individual recalcitrant lesions was often continued once weekly for periods of up to 36 weeks duration (mean 14.2 weeks). Patients received between one and three courses (average 1.7 courses; table 1). Typically, the initial courses were the longest (mean duration 11.6 weeks), while the mean duration of second treatment courses (in 10 patients) was 4.1 weeks. A third course (mean duration 2.3 weeks) was required in only four patients. Details of the doses and schedule of topical mustine were available for 10 patients (18 courses of mustine). On average, a total of 375.9 mg/m² (range 104.3–1148.1) was applied. The dose correlated with the extent of skin involvement. After initiation of mustine treatment, all except four patients (patients 6, 10, 11, and 18) were able to discontinue systemic steroid and/or cytostatic treatment. Figure 1 shows patient 19 before and one year after topical mustine treatment.

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Patient 19 before (top) and 1 year after (bottom) topical mustine therapy.

Topical mustine was well tolerated in 18 patients; two children developed irritant dermatitis caused by the treatment and characterised by non-pruritic erythema and oedema. The cutaneous irritation regressed on discontinuation of mustine, and did not recur when it was restarted. In two children with severe multisystem LCH, chronic skin disease was unresponsive to mustine (patients 6 and 18), and one of them subsequently died (table 2).

FOLLOW UP INVESTIGATIONS

Scars confined to areas of formerly active skin disease were found in six patients (30%). A similar percentage can be observed in children with cutaneous LCH not treated with mustine (unpublished data). No premalignant or malignant skin lesions were detected. Outcome evaluation revealed non-active and active disease, respectively, in 50% of the patients. Among the children with active disease, six had skin involvement. This was characterised by a mild scalp rash, resembling seborrhoeic dermatitis, that tended to improve gradually; it was classified as regressing (not requiring treatment) in five, and stable (requiring intermittent treatment) in one. Four children had progressive disease, two of them with skin lesions unresponsive to mustine treatment. Residual defects in other organ systems were noted in 12 patients (table 2).

Discussion

Topical mustine has been used in children with LCH skin involvement since 1979,8 but as yet, no study has systematically addressed the intermediate or long term outcome. Our follow up study indicates that topical mustine is a safe and effective treatment for severe skin lesions in most children with LCH. Skin disease healed either completely or was greatly improved in 18/20 patients within 21 days of mustine treatment, but—in accordance with previous reports9 26—50% of patients required two or three courses of treatment. Clearance of LCH skin lesions is more rapid and more lasting than in patients with mycosis fungoides.26 27 In 80% of our patients, no other systemic or topical drug was required to control skin lesions.

In single system LCH (usually bone, sometimes skin involvement alone), with a high rate of spontaneous regression, the basic question is whether or not treatment is needed.28 This dilemma is particularly difficult in “pure cutaneous” LCH.3 4However, cutaneous LCH lesions frequently do not respond to topical agents such as steroid ointment. In case of multisystem involvement, control is usually gained with corticosteroids or chemotherapy (high dose methylprednisolone, vinblastine, or etoposide1 2 24), but etoposide can cause leukaemia29 and is usually avoided if the overall prognosis is favourable. Radiation of indolent skin lesions has been abandoned, particularly in children. PUVA–chemotherapy carries an increased risk of skin cancer including malignant melanoma.30

Despite its proven efficacy, topical mustine is still not widely used for treatment of severe skin disease in LCH, largely because of lingering concern about tumourigenicity and the recognised local irritant effect of the drug in adults.12-21 Possible induction of skin cancer has been a subject of major concern since the introduction of mustine treatment for mycosis fungoides (in 1956), psoriasis (in 1959), and LCH (in 1975).8 26 27 After a latency period of two to seven years following topical mustard, increased rates (two- to ninefold) of epithelial cancers (basal cell carcinoma, squamous cell carcinoma, actinic keratoses, or keratoakanthoma) were reported in patients both with mycosis fungoides12 13 15 17 18 and psoriasis.14 16 Because of a previous report suggesting that ultraviolet B photocarcinogenesis was increased by mustine,11 sun protection seems to be advisable during and after treatment, but skin tumours were also reported in areas not exposed to the sun, and both in areas with and without previous mustine treatment. However, as pointed out previously,9 there are several confounding factors that make it difficult to attribute premalignant and malignant skin diseases in those reports to mustine alone.

Firstly, most of these patients had also received other potentially carcinogenic treatments (ultraviolet therapy, electron beam therapy, cytostatic drugs, or arsenic). Secondly, large areas of their bodies (up to 95% compared to 16% in our LCH patients) were exposed to topical mustine, often for much longer periods of time (on average, 81 months, and up to 12 years compared to 14 weeks—of mostly once weekly applications—in our patients), and at cumulative doses that were approximately 10–100-fold higher23 than in our patients. Finally, with a mean age of more than 50 years, the patients with psoriasis or mycosis fungoides had a much higher chance of spontaneously developing epithelial cancers. Interestingly, Studstrupet al were unable to detect DNA damage (sister chromatid exchanges) in lymphocytes after in vitro exposure to topical mustine.22 After a mean follow up period of 8.3 years, we found no evidence of serious side effects of topical mustine treatment. Specifically, no malignant or premalignant skin lesions were noted in either treated or untreated areas. The follow up period in our study was sufficiently long to cover the previously described latency period between exposure to mustine and development of skin cancers.12-18 For the moment, our results are reassuring, but they do not, of course, rule out the development of skin cancer in later years, and clinical follow up will be continued.

The rate of local reactions to mustine was low in our patients. Irritant dermatitis was observed in only two children (10%) and discontinuation of mustine was not required. The frequency of irritant or allergic contact dermatitis caused by mustine has been reported to be as high as 30–70% in adults,26 27 31 while others observed it in only 15% of their patients.21 It is often difficult to differentiate between irritant/toxic and allergic reactions to mustine.21 As with its carcinogenic effects, the occurrence of irritative contact dermatitis associated with mustine is probably related to the duration and extent of its application,27 thereby explaining the discrepant reported rates of local reactions. Several effective methods for desensitisation have been described.26 27 Other cutaneous side effects include transient hyperpigmentation6 27 and hypopigmentation. We found hyperpigmentation in one patient. Scars (confined to areas of former severe skin involvement) were noted in in six of 20 patients (30%), but such changes can be observed in LCH patients with severe skin disease irrespective of therapy,3 4 32 and are therefore considered a consequence of the disease, not the treatment.

In summary, topical mustine is a reasonable treatment option when skin disease is severe and systemic treatment is either unwarranted or ineffective. Preventive safety measures for patients (focal application, followed by thorough cleansing of the skin, subsequent sun protection) and for parents/personnel (gloves, face masks, gowns, hospital rooms specifically assigned for the administration of mustine33) must be strictly observed. Provided these precautions are taken, we and others2 6 9 10 22 23 27 regard the actual risks of topical mustine treatment as acceptable. The principal benefit is a 90% chance of an excellent short and medium term response, with few side effects, in a condition that is at best uncomfortable and at worst life threatening.