Sign of Leser-Trélat associated with adenocarcinoma of the rectum

ARTICLE

The sign of Leser-Trélat (SLT) is usually defined as the abrupt appearance and/or rapid increase in size and number of multiple seborrheic keratoses (SK) in association with an underlying cancer. Although these SK may show some characteristic features (distribution on the trunk, small size, and macular morphology), the most important criterium in diagnosing SLT is the dynamics of the eruption (acute onset and rapid growth of the lesions) [1]. Associated malignancies are, in order of frequency, gastrointestinal adenocarcinomas, hematology cancers, and breast and lung cancers, although a wide variation of malignances has been reported [2]. SLT implies a poor prognosis, because the average survival time after cancer diagnosis is about 11.5 months [1]. The existence of this entity is still controversial [3]. The literature and clinical practice bring contradictory data that do not clear up the issue definitely.

In this paper, we report a patient who met the classical features of SLT, which allowed us to diagnose a rectal adenocarcinoma at a curable stage.

Case report

A 75-year-old man presented in our department with multiple cutaneous lesions that had started appearing 3 months before. He described the lesions as "spots" and they were slightly pruriginous. His past medical history included benign prostatic hypertrophy and arterial hypertension treated with enalapril and hydrochlorothiazide. He had undergone appendectomy thirty years previously. He had no known allergies. His general health status had not changed during recent months. On examination we observed a large number of flat tumors sharply demarcated, yellowish to brown in color, some with a verrucous surface and greasy appearance, mainly located on the trunk (back, chest and abdomen), but also on the arms and thighs, clinically very suggestive of SK (Fig. 1). There were isolated, small vascular tumors (senile angiomas) on the trunk. No other cutaneous lesions were detected. A biopsy specimen from one of the lesions showed an epidermal tumor composed of basaloid cells, with acanthosis, papillomatosis, and marked hyperkeratosis. Pseudohorn cyst formation was frequently observed (Fig. 2). A diagnosis of SLT was made and then a complete anamnesis and physical examination were accomplished in order to detect an underlying malignancy, but abnormalities were not observed at this stage. Complete blood count was normal: hemoglobin 13.5 g/dL (normal range 13-16.5), platelets 360 x 10⁹/L (normal range 150-450 x 10⁹/L), total white cell count 7.0 x 10⁹/L (normal range 4-10.5 x 10⁹/L), neutrophils 3.9 x 10⁹/L (normal range 2.1- 4.9 x 10⁹/L), and lymphocytes 2.0 x 10⁹/L (normal range 1.4-2.9 x 10⁹/L). Serum biochemistry included urea, creatinine, glucose, albumine, uric acid, bilirrubin, AST, ALT, HDL and LDL cholesterol, triglycerides, and electrolites (sodium, potassium, and calcium). All of these were normal, except a moderate hyperglycemia of 117 mg/dL (normal range 74-105 mg/dL). The urinalysis was normal. Chest radiograph and abdominal ultrasonography did not reveal any pathological feature. Carcinoembryonic antigen levels were slightly elevated (6.2 UI/mL, normal range 0-2.5 UI/mL) and tissue polypeptide specific antigen (TPS) levels were also elevated (187 UI/mL, normal range 0-80 UI/mL), but the levels of other tumor markers (PSA, NSE, Ca 19.9, Ca 125, and alpha-fetoprotein) were within normal ranges. A test for occult blood in feces (modified guaiac method) was positive. Urologic (physical exploration and endorectal sonography), pulmonary (bronchoscopy), otorhinolaryngologic (indirect laryngoscopy) and, specially, digestive investigations (double-contrast upper gastrointestinal radiology, upper digestive endoscopy, barium contrast enema, and colonoscopy) were carried out. They revealed that the patient had a moderately differentiated adenocarcinoma of the rectum (Dukes A stage) (Fig. 3). The patient is asymptomatic after a follow-up of 2 years, but his SK remain unmodified.

Discussion

SLT consists of the association of the abrupt appearance and/or rapid increase in size and number of multiple SK and an underlying malignancy. Although there are a high number of case-reports in the literature, the existence of this sign is questioned by several authors [3-5]. The detractors of the existence of the SLT state that SK and cancer are extraordinarily common in elderly persons, so that it seems that both are determined by age. Although the average age at onset of SLT is about 60 years, young patients with SLT have been reported [6, 7], which would invalidate age as a determining factor for the appearance of both conditions, at least in these case-reports.

The abundance of case-reports in the literature may argue in favour of the existence of SLT [1]. In 1996, Swartz [2] counted 86 patients with this diagnosis. Nevertheless, the analysis of the literature is biased because the reported articles include only patients with malignancy and not people with eruptive SK without cancer. In fact, when studies based on patient series are considered, no evidence of association of SK and cancer is found [4, 5, 8].

Sixty-five per cent of the patients suffering SLT have concomitantly other paraneoplastic disorders (the most frequent is acanthosis nigricans, in one third of cases), which supports the existence of this sign [1]. Moreover, SLT shares some features with florid cutaneous papillomatosis, an obliged paraneoplastic dermatosis: there are patients with mixed forms, having both types of lesions (SK and papillomas) and both dermatoses are associated with acanthosis nigricans and the same type of malignancies [3, 9-11]. A specific characteristic of the well-defined paraneoplastic dermatoses that, nonetheless, SLT does not fullfil, is its parallelism with the course of the malignancy, since it only happens in one third of case-reports [1, 3].

Paraneoplastic dermatoses characterized by cellular proliferation, such as acanthosis nigricans and florid cutaneous papillomatosis, are thought to be due to growth-factors production by malignant cells, though this has been only documented in some isolated case-reports [12, 13]. The growth hormone has been implied in SLT pathogenesis [12, 14, 15], which goes along with the detection of growth hormone receptor in SK [16]. Other growth factors implied have been the alpha-transforming growth factor [13] and the epidermal growth factor [14]. This possible common pathogenic mechanism mediated by growth factors also supports the existence of SLT.

Although SK in SLT are usually small and macular, and are located on the trunk, they do not show any clinical nor pathological feature that definitively differentiates them from common SK. For this reason, the most important criterion in diagnosing SLT is the dynamics of the eruption. However, an exact definition of SLT, especially about the number and/or the rhythm of appearance of the SK, is lacking [1]. Therefore, the clue for the diagnosis relies on the history of the eruption provided by the patient, whose reliability is relative. Rampen and Schwengle [3] have shown that there is a statistically significant trend towards a shorter history of SK onset with advancing age, which is attributed to the unreliability of anamnestic data in elderly persons. Nevertheless, there is no method to diagnose SLT other than these anamnestic data.

CONCLUSION

In summary, from the available data it is not possible either to demonstrate or to reject definitively the existence of SLT. Therefore, if a patient gives us a history of acute onset and/or rapid increase in size and number of multiple SK, we must carry out studies in order to rule out an underlying cancer, that may still be at a curable stage. Our case-report is an illustration of this statement. The rectal adenocarcinoma would not have been detected so early if the patient had not consulted for the sudden appearance of his SK, since there had been no symptoms suggestive of that malignancy.

The best way to achieve a definitive conclusion about the existence or non-existence of SLT would be to study the incidence of malignant neoplasms in patients with clinical pictures suggestive of SLT, and compare it with the incidence of cancer among the general population. As SLT is an infrequent disorder, it would be necessary to investigate patients from many centers, but the first requirement would be to set convenient diagnostic criteria. Until this is done, controversy will remain.

Article accepted on 15/2/01

REFERENCES

1. Holdiness MR. On the classification of the sign of Leser-Trelat. J Am Acad Dermatol 1988; 19: 754-7.

2. Schwartz RA. Sign of Leser-Trélat. J Am Acad Dermatol 1996; 35: 88-95.

3. Rampen FHJ, Schwengle LEM. The sign of Leser-Trélat: does it exist? J Am Acad Dermatol 1989; 21: 50-5.

4. Grob JJ, Rava MC, Gouvernet J, Fuentes P, Piana L, Gamerre M, Sarles JC, Bonerandi JJ. The relation between seborrheic keratoses and malignant solid tumours. A case-control study. Acta Derm Venereol (Stockh) 1991; 71: 166-9.

5. Lindelöf B, Sigurgeirsson B, Melander S. Seborrheic keratoses and cancer. J Am Acad Dermatol 1992; 26: 947-50.

6. Westrom DR, Berger TG. The sign of Leser-Trélat in a young man. Arch Dermatol 1986; 122: 1356-7.

7. Barron LA, Prendiville JS. The sign of Leser-Trélat in a young woman with osteogenic sarcoma. J Am Acad Dermatol 1992; 26: 344-7.

8. Schwengle LEM, Rampen FHJ, Wobbes TH. Seborrhoeic keratoses and internal malignancies. A case control study. Clin Exp Dermatol 1988; 13: 177-9.

9. Yeh JSM, Munn SE, Plunkett, Harper PG, Hopster DJ, du Vivier AW. Coexistence of acanthosis nigricans and the sign of Leser-Trélat in a patient with gastric adenocarcinoma: a case report and literature review. J Am Acad Dermatol 2000; 42: 357-62.

10. Schwartz RA. Acanthosis nigricans, florid cutaneous papillomatosis, and the sign of Leser-Trélat. Cutis 1981; 28: 319-28.

11. Ginarte M, Sánchez-Aguilar D, Valdés F, Peteiro C, Toribio J. Acantosis nigricans, acantosis palmar y papilomatosis cutánea florida asociada a adenocarcinoma de cuello uterino. Actas Dermosifiliogr 1999; 90: 37-42.

12. Millard LG, Gould DJ. Hyperkeratosis of the palms and soles associated with internal malignancy and elevated levels of immunoreactive human growth hormone. Clin Exp Dermatol 1976; 1: 363-8.

13. Ellis DL, Kafka SP, Chow JC, Nanney LB, Inman WH, McCadden ME, King LE, et al. Melanoma, growth factors, acanthosis nigricans, the sign of Leser-Trélat, and multiple acrochordons: a possible role for alpha-transforming growth factor in cutaneous paraneoplastic syndromes. N Engl J Med 1987; 317: 1582-7.

14. Curry SS, King LE. The sign of Leser-Trélat. Arch Dermatol 1980; 116: 1059-60.

15. Klimer SL, Berman B, Morhenn VB. Eruptive seborrheic keratoses in a young man with acromegaly. J Am Acad Dermatol 1990; 23: 991-4.

16. Ginarte M, García-Caballero T, Fernández-Redondo V, Beiras A, Toribio J. Expression of growth hormone receptor in benign and malignant cutaneous proliferative entities. J Cutan Pathol 2000; 27: 276-82.