HEMATOLOGICALONCOLOGY, VOL. 10, 141-147 (1992)
ANGIOCENTRIC T-CELL LYMPHOMA OF THE NOSE, PARANASAL SINUSES AND HARD PALATE AGUSTIN AVILES, LETICIA RODRIGUEZ*, RENALDO GUZMAN, ALEJANDFU TALAVERA, EDNA L. GARCIA
AND JOSE C. DIAZ-MAQUEO
Department of Hematology and * Pathology, Oncology Hospital, National Medical Center, I.M.S.S., Mexico, D.F. Mexico
SUMMARY Sixty-five cases of malignant lymphoma of the nose, paranasal sinuses and hard palate were retrospectively analysed to identify the presence or absence of angiocentric lesions. We observed that the 23 patients with angiocentric lesions had a worse prognosis with a shorter duration of response and also a shorter duration of survival, compared with 42 cases of malignant lymphoma of the same anatomical region but without angiocentric lesions. Patients with angiocentric lymphoma were associated with other bad prognostic factors such as elevated levels of lactic dehydrogenase and beta 2 microglobulin, local bone destruction and lymphopenia. Immunophenotyping studies showed that most patients with angiocentric lesions had T cell lymphomas (18 of 23,78 per cent). We believe that patients with angiocentric T cell lymphomas of the nose, paranasal sinuses and hard palate represent a distinctive clinico-pathological entity with different clinical presentation and outcome. Patients with angiocentric T cell lymphomas had frequent relapse at extranodal sites and combined therapy should be considered as the initial therapeutic approach. KEY WORDS
Angiocentric T-cell lymphoma
Nose
Paranasal sinuses Hard palate
INTRODUCTION Jaffe’ coined the term ‘angiocentric immunoproliferative lesions’ to encompass a spectrum of lymphoproliferative disorders ranging from low-grade lesions of uncertain histogenesis to highgrade angiocentric lymphomas. This lesion generally has a poor prognosis with short survival and poor response to conventional treatment.” Malignant lymphomas of the head and neck region that originate in the nasal, cavity, paranasal sinuses and hard palate form an interesting and frequently diagnostically difficult group, since many cases are included in the clinical spectrum of the so-called non-healing ‘lethal midline granuloma syndrome’.s-’’The most effective treatment of this form of lymphoma remains unclear. Variable results had been reported, probably because lymphomas involving this region include a widely heterogeneous group of lesions.I2 ~ ’ no ~ ~close Angiocentric lesions had been described in this anatomical region in some c a ~ e s ,but correlation has been found between pathological findings and clinical c o ~ r s e . ’ - ’ ~ - ~ ~ In this study we have addressed the question of the utility of the presence or absence of angiocentric lesions as a guide to clinical management in patients with malignant lymphoma of the nose, paranasal sinuses and hard palate. Addressee for correspondence: Agustin Aviles, Apartado Postal 7-1220 06700, Mexico, D.F. Mexico.
02784232/92/03014147 $08.50 0 1992 by John Wiley & Sons, Ltd.
Received29 July 1991 Revised 14 April 1992
I42
A. AVILES ET AL.
MATERIALS AND METHODS Sixty-five cases were diagnosed as lymphoma of the nose, paranasal sinuses and hard palate for a three-year period (January 1980 to December 1982) in the Oncology Hospital. National Medical Center. Inclusion criteria were as follows: (a) pathologic diagnosis of lymphoma according to the criteria of the International Working Formulation;” (b) localized disease limited to the nose, paranasal sinuses, and hard palate with nodal or extranodal infiltration (local bone destruction was considered part of the same lesion); (c) presence of angiocentric lesions (angiocentric and angiodestructive lymphoid proliferation of small arteries and veins with a monomorphism of the infiltrate and marked cytologic atypia in both, small and large lymphoid cells) (Figure 1). An inflammatory background was usually inconspicuous but necrosis was p r ~ m i n e n t . ~ Patients underwent the following staging tests: complete blood count with differential and platelets, serum chemistry panel, serum protein electrophoresis, urinalysis, serum lactic dehidrogenase (LDH) and beta 2 microglobulin; chest X-ray, computerized tomography (CT) of abdomen and paranasal region; barium studies and upper endoscopy with gastric biopsies; bone marrow biopsy and smear were done in all patients. Sixty patients had serologic examination for human immunodeficiency virus (HIV). Immunohistochemical studies were performed on paraffin sections by the avidin-biotin peroxidase complex technique using a panel of monoclonal antibodies. Antibodies employed in the study with their dilutions were as follows: L26 ( x 100, Dakopatts, Denmark), MBl ( x 80, Bio-Science Products, Switzerland); LN1 (prediluted, Techniclone Int. Co., USA), LN2 (prediluted, Techniclone), MTl ( x 25, Bio-Science Products) and UCHLl ( x 25, Dakopatts). The L26, MBI, LN1 and LN2 antibodies were used as B-cell markers while the MTl and UCHLl antibodies were used as T-cell markers. In addition to staging studies, the following clinical parameters were recorded: age, sex, presenting symptoms; local bone destruction. The patients were treated with radiotherapy as follows: irradiation was administered with a 6 MeV linear accelerator at the rate of 750-1250 cGy per week with a total dose of 4500 to 5000 cGy. The areas clinically involved, the adjacent lymphoid region and Waldeyer’s ring were treated. No further treatment was given. Survival and relapse-free survival (RFS) were calculated by the Kaplan-Meier actuarial method. l 6 Differences between groups were analysed by using the generalized Wilcoxon test of Gehan.” Prognostic factors were studied using the multivariate regression technique of Cox.’* RESULTS Table 1 demonstrates that the clinical characteristics, B symptoms, high levels of LDH and beta 2 microglobulin, lymphopenia and local bone destruction were more frequent in patients with angiocentric lesions. Immunophenotyping studies showed that 18 out of 23 (78 per cent) marked as T cells. HIV studies were negative in all cases tested. Age, sex, histology did not reveal differences. Figure 2 show relapse-free survivals. Patients with angiocentric lymphomas had a median RFS of 16 months (range of 38 months). Patients with malignant lymphoma but without angiocentric lesions had a median RFS of 44 months (range 16 to 67 months). Statistically significant differences were evident (p < 0.001). All patients with angiocentric lesions have relapsed. The anatomical sites of relapse are seen in Table 2. Extranodal sites were most common in patients with angiocentric lesions. Figure 3 shows the duration of overall survival. Statistical differences were observed. Patients with angiocentric lesions had a median duration of survival of 25 months (range 6 to 76 months).
143
ANGIOCENTRIC T-CELL LYMPHOMA
Table 1. Clinical and laboratory characteristics Angiocentric lesion Yes No YO Number Number Total Age (years) median Sex: male female Local bone destruction B symptoms Lactic dehydrogenase (> 275 UI/L) Hemoglobin (< 12 g/dL) Lymphocytes (< 1.0 x 109/L) Histology: Diffuse large cell Immunoblastic Diffuse mixed Beta microglobulin ( > 3.5 pg/ml) T cell immunophenotype B cell immunophenotype Not done
23 35 13 10 16 19 16 5 10
100
16 6
69 29
56
44 69 82 69 21 43
1
5
14 18
60 78
1
4
4
17
%
42 38 26 16 10 8 8 0 0
100
30 10 2 4 6 30 6
71 23 6 9 14 71 14
61 39 23 19 19 0 0
Figure 1. Biopsy of paranasal lymphoma. The angiocentric lesion is seen as well as infiltration by the large cell lymphoma, H and E, 40 x
Patients with malignant lymphomas, but without angiocentric lesions remain alive and free of disease after they were treated with intensive chemotherapy (range of 20 to 93 +months [ p
ANGIOCENTRIC T-CELL LYMPHOMA OF THE NOSE, PARANASAL SINUSES AND HARD PALATE AGUSTIN AVILES, LETICIA RODRIGUEZ*, RENALDO GUZMAN, ALEJANDFU TALAVERA, EDNA L. GARCIA
AND JOSE C. DIAZ-MAQUEO
Department of Hematology and * Pathology, Oncology Hospital, National Medical Center, I.M.S.S., Mexico, D.F. Mexico
SUMMARY Sixty-five cases of malignant lymphoma of the nose, paranasal sinuses and hard palate were retrospectively analysed to identify the presence or absence of angiocentric lesions. We observed that the 23 patients with angiocentric lesions had a worse prognosis with a shorter duration of response and also a shorter duration of survival, compared with 42 cases of malignant lymphoma of the same anatomical region but without angiocentric lesions. Patients with angiocentric lymphoma were associated with other bad prognostic factors such as elevated levels of lactic dehydrogenase and beta 2 microglobulin, local bone destruction and lymphopenia. Immunophenotyping studies showed that most patients with angiocentric lesions had T cell lymphomas (18 of 23,78 per cent). We believe that patients with angiocentric T cell lymphomas of the nose, paranasal sinuses and hard palate represent a distinctive clinico-pathological entity with different clinical presentation and outcome. Patients with angiocentric T cell lymphomas had frequent relapse at extranodal sites and combined therapy should be considered as the initial therapeutic approach. KEY WORDS
Angiocentric T-cell lymphoma
Nose
Paranasal sinuses Hard palate
INTRODUCTION Jaffe’ coined the term ‘angiocentric immunoproliferative lesions’ to encompass a spectrum of lymphoproliferative disorders ranging from low-grade lesions of uncertain histogenesis to highgrade angiocentric lymphomas. This lesion generally has a poor prognosis with short survival and poor response to conventional treatment.” Malignant lymphomas of the head and neck region that originate in the nasal, cavity, paranasal sinuses and hard palate form an interesting and frequently diagnostically difficult group, since many cases are included in the clinical spectrum of the so-called non-healing ‘lethal midline granuloma syndrome’.s-’’The most effective treatment of this form of lymphoma remains unclear. Variable results had been reported, probably because lymphomas involving this region include a widely heterogeneous group of lesions.I2 ~ ’ no ~ ~close Angiocentric lesions had been described in this anatomical region in some c a ~ e s ,but correlation has been found between pathological findings and clinical c o ~ r s e . ’ - ’ ~ - ~ ~ In this study we have addressed the question of the utility of the presence or absence of angiocentric lesions as a guide to clinical management in patients with malignant lymphoma of the nose, paranasal sinuses and hard palate. Addressee for correspondence: Agustin Aviles, Apartado Postal 7-1220 06700, Mexico, D.F. Mexico.
02784232/92/03014147 $08.50 0 1992 by John Wiley & Sons, Ltd.
Received29 July 1991 Revised 14 April 1992
I42
A. AVILES ET AL.
MATERIALS AND METHODS Sixty-five cases were diagnosed as lymphoma of the nose, paranasal sinuses and hard palate for a three-year period (January 1980 to December 1982) in the Oncology Hospital. National Medical Center. Inclusion criteria were as follows: (a) pathologic diagnosis of lymphoma according to the criteria of the International Working Formulation;” (b) localized disease limited to the nose, paranasal sinuses, and hard palate with nodal or extranodal infiltration (local bone destruction was considered part of the same lesion); (c) presence of angiocentric lesions (angiocentric and angiodestructive lymphoid proliferation of small arteries and veins with a monomorphism of the infiltrate and marked cytologic atypia in both, small and large lymphoid cells) (Figure 1). An inflammatory background was usually inconspicuous but necrosis was p r ~ m i n e n t . ~ Patients underwent the following staging tests: complete blood count with differential and platelets, serum chemistry panel, serum protein electrophoresis, urinalysis, serum lactic dehidrogenase (LDH) and beta 2 microglobulin; chest X-ray, computerized tomography (CT) of abdomen and paranasal region; barium studies and upper endoscopy with gastric biopsies; bone marrow biopsy and smear were done in all patients. Sixty patients had serologic examination for human immunodeficiency virus (HIV). Immunohistochemical studies were performed on paraffin sections by the avidin-biotin peroxidase complex technique using a panel of monoclonal antibodies. Antibodies employed in the study with their dilutions were as follows: L26 ( x 100, Dakopatts, Denmark), MBl ( x 80, Bio-Science Products, Switzerland); LN1 (prediluted, Techniclone Int. Co., USA), LN2 (prediluted, Techniclone), MTl ( x 25, Bio-Science Products) and UCHLl ( x 25, Dakopatts). The L26, MBI, LN1 and LN2 antibodies were used as B-cell markers while the MTl and UCHLl antibodies were used as T-cell markers. In addition to staging studies, the following clinical parameters were recorded: age, sex, presenting symptoms; local bone destruction. The patients were treated with radiotherapy as follows: irradiation was administered with a 6 MeV linear accelerator at the rate of 750-1250 cGy per week with a total dose of 4500 to 5000 cGy. The areas clinically involved, the adjacent lymphoid region and Waldeyer’s ring were treated. No further treatment was given. Survival and relapse-free survival (RFS) were calculated by the Kaplan-Meier actuarial method. l 6 Differences between groups were analysed by using the generalized Wilcoxon test of Gehan.” Prognostic factors were studied using the multivariate regression technique of Cox.’* RESULTS Table 1 demonstrates that the clinical characteristics, B symptoms, high levels of LDH and beta 2 microglobulin, lymphopenia and local bone destruction were more frequent in patients with angiocentric lesions. Immunophenotyping studies showed that 18 out of 23 (78 per cent) marked as T cells. HIV studies were negative in all cases tested. Age, sex, histology did not reveal differences. Figure 2 show relapse-free survivals. Patients with angiocentric lymphomas had a median RFS of 16 months (range of 38 months). Patients with malignant lymphoma but without angiocentric lesions had a median RFS of 44 months (range 16 to 67 months). Statistically significant differences were evident (p < 0.001). All patients with angiocentric lesions have relapsed. The anatomical sites of relapse are seen in Table 2. Extranodal sites were most common in patients with angiocentric lesions. Figure 3 shows the duration of overall survival. Statistical differences were observed. Patients with angiocentric lesions had a median duration of survival of 25 months (range 6 to 76 months).
143
ANGIOCENTRIC T-CELL LYMPHOMA
Table 1. Clinical and laboratory characteristics Angiocentric lesion Yes No YO Number Number Total Age (years) median Sex: male female Local bone destruction B symptoms Lactic dehydrogenase (> 275 UI/L) Hemoglobin (< 12 g/dL) Lymphocytes (< 1.0 x 109/L) Histology: Diffuse large cell Immunoblastic Diffuse mixed Beta microglobulin ( > 3.5 pg/ml) T cell immunophenotype B cell immunophenotype Not done
23 35 13 10 16 19 16 5 10
100
16 6
69 29
56
44 69 82 69 21 43
1
5
14 18
60 78
1
4
4
17
%
42 38 26 16 10 8 8 0 0
100
30 10 2 4 6 30 6
71 23 6 9 14 71 14
61 39 23 19 19 0 0
Figure 1. Biopsy of paranasal lymphoma. The angiocentric lesion is seen as well as infiltration by the large cell lymphoma, H and E, 40 x
Patients with malignant lymphomas, but without angiocentric lesions remain alive and free of disease after they were treated with intensive chemotherapy (range of 20 to 93 +months [ p
