
	Version PDF

Routine bone marrow biopsy in the initial evaluation of primary cutaneous B-cell lymphoma does not appear justified

European Journal of Dermatology. Volume 19, Number 3, 216-20, May-June 2009, Investigative report

DOI : 10.1684/ejd.2009.0633

Summary

Author(s) : Gaëlle Quereux, Anne Sophie Frot, Anabelle Brocard, Cécile Leux, Jean-Jacques Renaut, Brigitte Dreno , Dermatology Department, Nantes University Hospital Center, Place Alexis Ricordeau, 44093 Nantes, France, Department of Biostatistics, Nantes University Hospital Center.

Summary : Primary cutaneous B-cell lymphomas are a rare entity. They are included in the distinct classification of primary cutaneous lymphomas, the WHO-EORTC. In order to be confirmed, the primary nature of a cutaneous lymphoma requires that negative results of a CT scan of the chest, abdomen and pelvis and of a bone marrow biopsy (BMB) be obtained. Nevertheless, there is a question as to whether BMB should be performed routinely in view of the good prognosis of certain cutaneous B-cell lymphomas and the invasive nature of the examination. To answer that question, we studied retrospectively 62 cases of cutaneous B-cell lymphomas in which a BMB was performed. In 4 cases, lymph nodes, and in one case pancytopenia were identified during the initial evaluation performed at the same time as BMB and thus these patients were excluded from the analysis. Among the 57 patients, the BMB was positive in only 3 patients (5.2%). Interestingly, the positivity of the biopsy did not significantly affect the way that the treatment was given. In conclusion, this study demonstrates that it is not indispensable to perform a routine BMB for a Primary cutaneous B-cell lymphoma with cutaneous lesions on examination and with negative CT Scan and blood laboratory evaluations.

Keywords : primary cutaneous B-cell lymphoma, bone marrow biopsy

Pictures

ARTICLE

Auteur(s) : Gaëlle Quereux¹, Anne Sophie Frot¹, Anabelle Brocard², Cécile Leux², Jean-Jacques Renaut¹, Brigitte Dreno²

¹Dermatology Department, Nantes University Hospital Center, Place Alexis Ricordeau, 44093 Nantes, France
²Department of Biostatistics, Nantes University Hospital Center

accepté le 10 Decembre 2008

Primary cutaneous (B-cell and T-cell) lymphomas are a group of rare diseases for which there are few epidemiological data. The incidence of these primary cutaneous lymphomas is reported to be in the order of 10 cases per million per year, approximately [1]. Cutaneous B-cell lymphomas account for about 25% of cutaneous lymphomas [2-4]. They are said to be primary when they are localized only in the skin with no systemic involvement (lymph node or bone marrow). The uniqueness of these primary cutaneous lymphomas therefore justified the creation of their own WHO/EORTC classification [5] in 2005. But the TNM system used for Mycosis Fungoides and Sezary syndrome was not appropriate for other primary cutaneous lymphomas. Thus a unified TNM classification system applicable for all primary cutaneous lymphomas has been recently created by both the International Society for Cutaneous Lymphomas (ISCL) and the cutaneous lymphoma task force of the European Organization of Research and Treatment of Cancer (EORTC) [6].

Conventionally, the term “primary cutaneous lymphoma” is applied only when the staging investigation is negative at the time that the lymphoma is diagnosed and remains negative for the next six months. This period of 6 months has been called into question by several authors [5], for two principal reasons: first, because it is difficult at the time of the diagnosis to wait 6 months (to make a definitive diagnosis) before offering the patient a treatment, and second, in the case of an aggressive form, certain cutaneous lymphomas can spread systemically in less than 6 months. However, since the prognosis of primary cutaneous lymphomas is often much better than that of systemic lymphomas, it is important to make a distinction between them, in order to avoid unnecessary aggressive treatments [3, 7]. The evaluation for arriving at a diagnosis of primary cutaneous B-cell lymphoma includes notably a CT scan of the chest, abdomen and pelvis and a bone marrow biopsy, which have to be negative. Nevertheless, the question of whether the BMB should be done routinely arises because of its low positivity in our daily practice, the good prognosis of most of these cutaneous B-cell lymphomas, and while this examination remains invasive and is often poorly accepted by the patient.

In this context, we carried out a retrospective study of cutaneous B-cell lymphomas revealed by cutaneous lesions in order to check the percentage of positive BMBs at the initial evaluation, the treatment chosen as a result, and thus determine whether or not that examination was justified.

Patient population and methods

This was a retrospective study of 62 patients followed in the Department. All the patients included presented with one or more cutaneous lesions revealing a cutaneous B-cell lymphoma, which was the reason for the consultation. The diagnosis of cutaneous B-cell lymphoma was confirmed by a histological examination classifying the lymphomas according to the WHO-EORTC classification. The following histological subtypes were noted: primary cutaneous follicle center lymphomas (PCFCL), primary cutaneous diffuse large B-cell lymphomas leg type (PCDLBCL) and primary cutaneous marginal zone B-cell lymphomas (PCMZBCL). The staging investigation included a BMB, a CT scan of the chest, abdomen and pelvis and a laboratory evaluation (CBC, lymphocytes in the circulating blood).

The patients included were followed from the time of the diagnosis until the last update (date of the last consultation or date of death).

The principal objective of this study was to determine the percentage of positive BMBs at the time of the initial evaluation. The secondary objectives were to compare the clinical, immunohistological and molecular profiles of these primary Cutaneous B-Cell Lymphomas, and in addition the relapse-free survival and overall survival of these cutaneous B-cell lymphomas. The clinical characteristics studied were age, sex, site of cutaneous involvement. The immunohistological criteria were the phenotypical expression of B lymphocytes (CD19, 20, 21, 22) studied by using an immunohistochemical technique (peroxydase) on a frozen section of cutaneous lesions. The rearrangement of gene B in the tumoral lesion was studied by a PRC technique (PCR FR1 and FR2, Biomed). BMB was considered as positive when two or more islets of atypical lymphocytes or a diffuse infiltrate of atypical B lymphocytes was identified, confirmed by a B phenotype and a double reading by 2 pathologists.

Statistical analysis

For the statistical analysis, the tests employed for the bivariate analyses were the following:

– To compare two means: the Student t test if its validity conditions were met, otherwise the Mann-Whitney non-parametric test.
– To compare more than two means: a one-factor analysis of variance.
– To compare two percentages or more: the chi-square test if the theoretical numbers were high enough, otherwise the Fisher test (two percentages) or a chi-square test with simulations of p (more than two percentages).
– To study the relationship between two quantitative variables: the null hypothesis test of the Pearson correlation coefficient if its conditions of validity were met, otherwise Spearman’s rank correlation test.

All the tests were performed with a type 1 error probability at the level of 5%.

We were unable to carry out a multi-criteria analysis because of insufficient numbers (in each subclass), particularly in the subpopulation with a positive BMB.

Results

62 patients were included in this study. But 4 presented with deep lymph nodes revealed with CT Scan performed at the same time as the BMB and 1 patient had peripheral pancytopenia, also discovered during this initial evaluation. Thus, they were excluded from the analysis and considered as systemic lymphomas with cutaneous localization.

Principal objective: percentage of positive BMB in primary Cutaneous B-Cell Lymphomas

The final analysis was performed on 57 cases. BMB was positive in 3/57 (5.2%) of primary Cutaneous B-Cell Lymphomas.

Here is a full report of these 3 cases: 1st case: This was a 61-year-old woman who had a single cutaneous lesion on her back, whose histological features were those of a cutaneous follicle center lymphoma. A routine BMB demonstrated a focal interstitial bone marrow infiltration. A decision was made of abstention of any treatment because of the absence of lymph node involvement, and of the normal blood count. After surgical excision of the single cutaneous localization and 29 months of follow-up, the patient was still in remission. 2nd case: This was a 53-year-old woman with several cutaneous lesions of the scalp, whose histological features were those of a cutaneous follicle center lymphoma. The routine BMB found 2 lymphoid islets. Because of the low density of the bone marrow infiltration, a decision was made, in agreement with the hematologists, of abstention of any treatment and follow up. The cutaneous lesion was treated by radiotherapy. A year later, the patient presented with a diffuse cutaneous relapse. Treatment with 4 injections of anti-CD20 antibodies (Mabthera^®) produced a remission for 6 years. 3rd case: This was a 34-year-old man who had several cutaneous lesions of the forehead (left temporal) and of the back (7 lesions), whose histological features were those of a cutaneous marginal zone B-cell lymphoma. The BMB found two wide paratrabecular lymphoid islets whose low tumor mass did not justify a specific treatment, according to the hematologists. The cutaneous lesions were treated by radiotherapy with a complete remission. The patient was still in cutaneous remission 6 years later.

Thus, in these 3 cases, the discovery of bone marrow invasion did not change the therapeutic management and the duration of remission.

Secondary objectives

Clinical, immunological, histological characteristics

Histological distribution of primary cutaneous B-cell lymphomas

Of the 57 PCBCL cases, the distribution consisted of 39 cutaneous follicle center lymphomas, 16 cutaneous diffuse large B-cell lymphomas, leg type, and 2 cutaneous marginal zone B-cell lymphomas (table 1)
Table 1 Number of cutaneous lesions and histological subtype

No. of patients	Single cutaneous lesion	Several cutaneous lesions
Cutaneous follicle center lymphomas	17	20
Diffuse large-cell B-cell lymphomas leg type.	9	7
Cutaneous marginal zone B-cell lymphomas	0	1

Age

The mean age was 56.11 years for the cutaneous follicle center lymphomas, 56 years for cutaneous marginal zone B-cell lymphomas and 70.63 years for the cutaneous diffuse large B-cell lymphomas leg type. The age at the time of diagnosis was significantly different depending on the histological subtype (p = 0.006).

Sex

The distribution was the following: cutaneous follicle center lymphomas: 22 women and 17 men, cutaneous diffuse large B-cell lymphomas, leg type: 8 women and 8 men, cutaneous marginal zone B-cell lymphomas: 1 woman and 1 man. The M/F sex ratio was not statistically different according to the histological subtype.

Localizations of cutaneous lesions

The cutaneous follicle center lymphomas were localized more often on the back and the buttocks in comparison with the other histological subtypes (p < 0.05).

Phenotypical analysis of B lymphocytes in skin biopsies

For all types of primary B-cell lymphomas, the B lymphocyte phenotype found most often was CD20+; CD22+. Antigen CD20 was the antigen most often expressed (96.5%), followed by CD22 (75.4%) and CD19 (20.4%) and finally CD21 (16.3%).

Rearrangement of gene B in the tumoral cutaneous lesion

So far as the PCBCL were concerned, it was positive in 78.3% of the cutaneous follicle center lymphomas (18/23), 75% of the cutaneous diffuse large B-cell lymphomas leg type (9/12), 100% of the cutaneous marginal zone B-cell lymphoma (2 cases).

Evolution of relapse-free survival (RFS), overall survival (OS)

For this analysis the 5 systemic B-Cell Lymphomas were included. The median follow-up was 53 months (6 months-13 years).

Overall survival

There were 8 deaths among the 62 patients. (12.9%): 2 in the group of systemic lymphomas (25%) and 6 in the PCBCL group (11.1%). These 6 patients all had a cutaneous diffuse large B-cell lymphomas leg type (figure 1).

Relapse-free survival

36/62 patients had a relapse of their lymphoma (58%). Among those with systemic B-cell lymphomas from the outset, 2 had a relapse. In the PCBCL group, 34/57 patients relapsed (59.6%), including 20/37 (54%) of the cutaneous follicle center lymphomas, 13/16 (81%) of the cutaneous diffuse large B-cell lymphomas, leg type, 1 relapse of the cutaneous marginal zone B-cell lymphoma. The median relapse-free survival was 22 months. The minimum was 47 days and the maximum was 9 years (figure 2). Despite a higher relapse rate in the patients with cutaneous diffuse large B-cell lymphomas, leg type, the statistical test did not show a significant difference in the relapse rate according to histological subtype (p = 0.13). Of the 36 patients who had a relapse, 31 patients/36 (86.1%) had only a cutaneous relapse, 2 patients/36 (5.6%) had both a cutaneous and lymph node relapse, 3 patients/36 (8.3%) had only a lymph node relapse. Taking as criterion “a single lesion or multiple lesions” at the time of the diagnosis, the patients had a better relapse-free survival rate when they had a single lesion than when they had multiple lesions (p < 0.01) (figure 3)

Discussion

Our study shows that a BMB performed at the initial evaluation of a primary cutaneous B-cell lymphoma, confirmed by normal CT Scan and blood analysis, found a bone marrow invasion in only 3/57 patients (5.2%). In two cases, it was a cutaneous follicle center lymphoma and in one case a cutaneous marginal zone B-cell lymphoma. In addition, the BMB is not more frequently positive in patients with cutaneous diffuse large B-cell lymphomas leg type, even if the prognosis is worse. Furthermore, an interesting fact is that that the discovery of this bone marrow invasion did not change the therapeutic management of these three patients, even after discussion with hematologists. In the literature, only two studies discuss the value of performing a routine BMB in the initial evaluation of cutaneous B-cell lymphoma. A French study [8] involving 23 cases of cutaneous follicle center lymphomas and immunocytomas found a BMB positive in only one case, which made it possible to diagnose a low-grade non-Hodgkin’s B-cell lymphoma. A second study by Yasukawa et al. [9] actually included 70% secondary cutaneous B-cell lymphomas. Of the 20 B-cell lymphomas with a cutaneous expression, 6 were “pure” cutaneous and the other 14 were systemic B-cell lymphomas with a secondary cutaneous expression (11 diffuse large-cell B-cell lymphomas, 2 follicle center lymphomas and 1 unclassifiable lymphoma). This last study did not state the number of positive BMBs.

ISCL/EORTC recommendations for staging in cutaneous lymphomas for the management of cutaneous B-cell lymphomas have recently been published. Bone marrow biopsy and aspirate are required in cutaneous diffuse large B-cell lymphomas and leg type, but considered as optional in cutaneous follicle center lymphoma and cutaneous marginal zone B-cell lymphoma, which is in agreement with our results except for cutaneous diffuse large B-cell lymphomas where our results show that bone marrow examination appears also to have a limited interest [10].

Concerning the age of primary CBCL patients, those with diffuse large-cell B-cell lymphomas, leg type, were appreciably older. These results agree with those found in the literature [11-13]. In the phenotypical analysis of the tumoral infiltrate, the tumor cells express mainly CD20+, CD22+ antigens, whatever the type of CPBCL. A B-cell clone is found in about 60% of primary cutaneous B-Cell lymphomas. These results are also comparable with the numbers usually found in the literature in respect to cutaneous B-cell lymphomas.

Concerning the relapse free survival, taking into account in the analysis the 5 systemic B-Cell lymphomas, the rate of relapse was 33% in these 5 systemic B-Cell lymphomas, with a median follow-up of 40 months. The relapses are cutaneous lesions and lymph nodes. It is 63% in primary cutaneous B-cell lymphomas with a median follow-up of 55 months. According to the histological subtype, it is 55% on follicle center lymphomas, 82% in diffuse large-cell B-cell lymphomas, leg type. The lack of statistical significance (p = 0.13) between the different subtypes is probably due to the fact that the size of the samples is too small. However, these rates of relapse appear higher than the ones reported in the study by Zinzani et al. [12] who found a rate of relapse of 46.7% for primary cutaneous B-cell lymphomas with a median follow-up of 12.5 years (46.5% of relapse for follicle center lymphomas, 44.4% for cutaneous marginal zone B-cell lymphomas, 54.8% for diffuse large-cell B-cell lymphomas leg type. This difference also was not significant. It should be noted that in our study, relapses are more frequent in all cases in the first two years. Besides, the site of the relapse is mostly cutaneous (86.1%). This figure is comparable to the one given by Zinzani et al. [12]. The risk of relapse in PCBCL appears to be related to the criterion “single lesion or multiple lesions” at the time of diagnosis, with a lower rate of relapse in the case of a single lesion (p < 0.01). Thus, the rate of relapse-free survival at 5 years was 60% when there was a single lesion and of only 20% when there were multiple lesions. This result was also found in the study by Zinzani et al. [12]. Furthermore, our study confirms the significantly less favorable evolution in the case of diffuse large-cell B-cell lymphomas, leg type, in comparison with the other histological subtypes, as was found among others by Willemze et al. [13].

Finally, with regard to the overall survival at 5 years, it is slightly lower for systemic versus primary lymphomas (73% vs 87%) but not significantly so. The results found regarding the PCBCL are comparable to those of two large retrospective studies [3, 12] in which the rates of survival of PCBCL at 5 years were 94% [12] and 89% [3].

In conclusion, our study shows that performing a routine BMB in the initial evaluation of a primary cutaneous B-cell lymphoma, whatever the histological subtype, with normal clinical examination, negative TC Scan and normal CBC and lymphocytes in blood, is not justified because of its low positivity but also because it does not significantly modify the treatment to be given.

Acknowledgements

Financial support: none. Conflict of interest: none.

Références

1 Saurat JH. Dermatologie et infections sexuellement transmissibles. 4^e édition. Masson.

2 Kerl H, Cerroni L. Primary cutaneous B-cell lymphomas: then and now. J Cutan Pathol 2006; 33 (Suppl.1): 1-5.

3 Willemze R, Kerl H, Sterry W, et al. EORTC classification for primary cutaneous lymphomas. Blood 1997; 90: 354-71.

4 Bouaziz JD, Bastuji-Garin S, Poszepczynska-Guigné E, Wechsler J, Bagot M. Relative frequency and survival of patients with primary cutaneous lymphomas: data from a single-centre study of 203 patients. Br J Dermatol 2006; 154: 1206-7.

5 Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas. Blood 2005; 105: 3768-85.

6 Kim YH, Willemze R, Pimpinelli N, Whittaker S, Olsen EA, Ranki A. Dummer, Hoppe RT TNM classification system for primary cutaneous lymphomas other than mycosis fungoides and Se’zary syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the Cutaneous Lymphoma Task Force of the European Organization of Research and Treatment of Cancer (EORTC). Blood 2007; 110: 2479-84.

7 Santucci M, Pimpinelli N, Arganini L. Primary cutaneous B-cell lymphoma: a unique type of low-grade lymphoma. Clinicopathologic and immunologic study of 83 cases. Cancer 1991; 67: 2311-26.

8 Wejroch MP, Cornillet P, Perceau G, Durlach A, Bernard P. Fréquence des cancers associés dans les lymphomes cutanés. Ann Dermatol Venereol 2004; 131: 339-45.

9 Yasukawa K, Kato N, Kodama K, Hamasaka A, Hatz H. The spectrum of cutaneous lymphomas in Japan: a study of 62 cases based on the WHO Classification. J Cutan Pathol 2006; 33: 487-91.

10 Senff NJ, Noordijk M, Kim YH, et al. European Organization for Research and Treatment of Cancer and International Society for Cutaneous Lymphoma consensus recommendations for the management of cutaneous B-cell lymphomas. Blood 2008; 112: 1600-9.

11 Fink-Puches R, Zenahlik P, Bäck B, Smolle J, Kerl H, Cerroni L. Primary cutaneous lymphomas: applicability of current classification schemes (EORTC, WHO) based on clinicopathologic features observed in a large group of patients. Blood 2002; 99: 800-5.

12 Zinzani PL, Quaglino P, Pimpinelli N, et al. Pronostic factors in primary cutaneous B-cell lymphoma: the Italian study group for cutaneous lymphomas. J Clin Oncol 2006; 24: 1376-82.

13 Hallermann C, Niermann C, Fischer RJ, Schulze HJ. New prognostic relevant factors in primary cutaneous diffuse large B-cell lymphomas. J Am Acad Dermatol 2007; 56: 588-97.