Patients with Classical Hodgkin Lymphoma with Less Than 10.5 CD20-Positive Reed–Sternberg Cells in 10 High-Power Fields Have Better Prognosis

Abstract

The CD20 immunoexpression in classical Hodgkin lymphoma (CHL) is variable and it may represent an intersection between the pathophysiology of CHL and the clinical prognosis of patients. Retrospective study with 174 patients with pathological diagnosis of CHL. Clinical data were reviewed from medical records. Immunophenotypic study with CD20 were performed in two ways: hotspot count and in 10 high-power fields (HPF). The data were tabulated and the p value less than 0.05 was considered significant. The CD20-positive phenotype was found in 45 (25.9%) of the total of 174 patients with available material. Patients with CD20 positivity on neoplastic cells tend to heal earlier, particularly those with less than 10.5 cells/10 HPF (p = 0.037). Patients with CHL with less than 10.5 CD20-positive Reed-Sternberg cells in 10 HPF have better prognosis.

Introduction

Typically, Hodgkin and Reed-Sternberg (HRS) cells express CD30, CD15 and PAX5 by immunohistochemistry (IHC), with CD20 expression frequently restricted to a subset of the tumor cell population with variable intensity in different parts of the cell membrane [1]. The positivity of CD20 in classical Hodgkin lymphoma (CHL) is reported in the literature with percentages ranging from 7.3 to 35.3% [2–11]. The differences observed in these studies may be related to interpretative criteria of the IHC, tissue sampling, therapies or may reflect geographic particularities. Although controversial, both in terms of the presence and extent of immunostaining, this positivity may represent an intersection between the pathophysiology of CHL and the clinical prognosis of patients.

Methods

We reviewed 174 tissue samples of patients from three Brazilian healthcare centers diagnosed with CHL between 1998 and 2019. The four histological subtypes were included: lymphocyte rich CHL (LRCHL), nodular sclerosis CHL (NSCHL), mixed cellularity CHL (MCCHL) and lymphocyte depleted CHL (LDCHL). Nodular lymphocyte predominant Hodgkin lymphoma (NLPHL) and grey zone cases were excluded. The Reed-Sternberg (RS) cells were evaluated by IHC with the CD20 marker (L26, Dako ready-to-use) in two ways: hotspot count and count in 10 high-power fields (HPF). Hotspot area was defined as visual identification of the spot with the highest expression of CD20 in × 400 magnification and the presence of at least one neoplastic cell expressing the marker was considered positive/hotspot. Ten HPF was defined as 10 fields of view at × 400 magnification around the hotspot area including it. Intensity assessment was not performed, as it was observed that the labeling of this antibody was weak and heterogeneous in all CHL cases of this study. The slides were seen by two pathologists and a consensus was made by them. Clinical aspects, including B symptoms, extranodal involvement, Ann Arbor staging, treatment with doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD regimen) and/or radiation therapy and/or bone marrow transplantation, bone marrow involvement or non-involvement, response to treatment (complete or progressive disease), recurrence and death were also collected. A poor outcome was considered for those patients with progressive disease, recurrence, or disease-related death in the last medical record.

Analyzes were performed using the statistical software SPSS for Windows, version 15.0 (SPSS Inc., Chicago, IL, USA), except for the survival curves that were performed in the statistical software R. The analysis was descriptive, followed by the Chi square test and Fisher’s exact test to compare proportions. The Kaplan–Meier method was used to estimate patient survival and the log-rank test used to compare survival curves between groups. Significance was determined using Student's t-test and statistical significance was achieved when the p value is < 0.05.

Results

Out of 174 patients, 129 (74.1%) were negative for CD20 in RS cells and 45 (25.9%) were positive for this marker. Table 1 shows the clinicopathological characteristics and Table 2 shows the morphological and immunohistochemical characteristics between the groups. The outcome differences observed between patients with CD20-positive neoplastic cells and those with CD20-negative neoplastic cells were not statistically significant (22.7% and 37.3%, p = 0.155, respectively) (Table 1). The median of neoplastic cells positive for CD20 in 10HPF was 10.5 (1–57). The disease-free survival was better in CD20-positive patients, particularly those with less than 10.5 cells/10HPF compared to patients with more than 10.5 cells/10HPF (p = 0.0037) (Fig. 1a). This is not seen for overall survival (p = 0.2310) (Fig. 1b). The median of neoplastic cells positive for CD20 in hotspot was 6 neoplastic cells (1–25). As to this hotspot count, CD20-positive patients had better overall survival and disease-free survival when compared to negative ones, but there was no statistical significance between them (p = 0.2539 and p = 0.0903, respectively). The differences between the positivity of neoplastic cells for CD20 and the other morphological and clinical data collected were not statistically significant (Tables 1, 2).

Table 1.

Clinicopathological characteristics of CD20-positive and CD20-negative patients with classical Hodgkin lymphoma

	CD20−	CD20+	p valuea
Sex (n = 174)
Male	77/129 (59.7%)	24/45 (53.3%)	0.284
Female	52/129 (40.3%)	21/45 (46.7%)
Age at diagnosis, y (n = 170)	30.5 (13–79.4)	28.6 (15–73.2)	0.613b
Location at diagnosis (n = 171)
Cervical	55/128 (43%)	29/43(67.4%)
Supraclavicular	23/128 (18.0%)	4/43 (9.3%)
Axillary	16/128 (12.5%)	4/43 (9.3%)
Inguinal	9/128 (7.0%)	2/43 (4.7%)	–
Mediastinal	16/128 (12.5%)	3/43 (7.0%)
Mediastinal (bulky)	7/128 (5.5%)	1/43 (2.3%)
Retroperitoneal	1/128 (0.8%)	0/43 (0.0%)
Spinal canal	1/128 (0.8%)	0/43 (0.0%)
Bulky disease (n = 108)	13/83 (15.7%)	2/25 (8.0%)	0.271
Extranodal involvement (n = 110)c	35/83 (42.2%)	10/27 (37.0%)	0.406
Bone marrow involvement (n = 141)	21/105 (20.0%)	5/36 (13.9%)	0.292
B symptoms (n = 109)	57/84 (67.9%)	16/25 (64.0%)	0.447
Ann Arbour stage (n = 108)
I—II	35/83 (42.2%)	14/25 (56.0%)	0.273
III—IV	48/83 (57.8%)	11/25 (44.0%)
HIV (n = 110)	3/85 (3.5%)	1/25 (4.0%)	0.649
First line treatment (n = 106)
No treatment	3/82 (3.7%)	0/24 (0.0%)	–
ABVD	74/82 (90.2%)	24/24(100%)
No ABVDd	5/82 (6.1%)	0/24 (0.0%)
Radiotherapy (n = 106)	43/82 (52.4%)	11/24 (45.8%)	0.368
Bone marrow transplant (n = 106)	8/82 (9.8%)	0/24 (0.0%)	0.118
Unfavorable outcome (n = 97)e	28/75 (37.3%)	5/22 (22.7%)	0.155
Histological subtype (n = 174)
Nodular sclerosis	93 (72.1%)	34 (75.6%)	0.042
Mixed cellularity	22 (17.1%)	5 (11.1%)
Lymphocyte-rich	2 (1.6%)	1 (2.2%)
Lymphocyte-depleted	2 (1.6%)	0 (0.0%)
Unclassifiable	10 (7.8%)	5 (11.1%)

^aChi-square test/Fisher's exact test (p value < 0.05)

^bMann–Whitney Test

^cBone marrow, spleen, liver, spinal canal, lung, small intestine, bone, breast, abdominal wall and skin

^dBEACOOP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine and prednisone), DHAP (dexamethasone, cisplatin and cytarabine), ICE (ifosfamide, carboplatin and etoposide) and GDP (gemcitabine, dexamethasone and cisplatin)

^eProgressive disease, relapse and death

ABVD (doxorubicin, bleomycin, vinblastine and dacarbazine)

Table 2.

Morphological and immunophenotypic characteristics between CD20-positive and CD20-negative groups

	CD20−	CD20+	Total	p valuea
Morphology
Maintenance of lymph node architecture	16 (16.2%)	3 (8.8%)	19 (14.3%)	0.225
Nodular pattern	57 (54.8%)	21 (61.8%)	78 (56.5%)	0.306
Sclerosis	75 (72.1%)	27 (79.4%)	102 (73.9%)	0.273
Capsular thickening	71 (68.3%)	22 (64.7%)	93 (67.4%)	0.426
Granuloma	11 (10.4%)	7 (20.6%)	18 (12.9%)	0.108
Eosinophilia	83 (78.3%)	24 (70.6%)	107 (76.4%)	0.242
Plasmacytosis	60 (56.6%)	20 (58.8%)	80 (57.1%)	0.491
Tingible body macrophages	29 (27.4%)	13 (38.2%)	42 (30.0%)	0.161
Endothelial reactivity	76 (71.7%)	26 (76.5%)	102 (72.9%)	0.380
Vascular ectasia	22 (24.2%)	8 (23.5%)	30 (24.0%)	0.571
Neoplastic cells (IHC e CISH)
CD30 dot pattern	80 (75.5%)	28 (82.4%)	108 (77.1%)	0.281
CD15	50 (47.2%)	18 (52.9%)	68 (48.6%)	0.349
PAX5	70 (71.4%)	27 (79.4%)	97 (73.5%)	0.251
LMP1	28 (28.0%)	7 (20.6%)	35 (26.1%)	0.270
EBER	20 (21.1%)	6 (17.6%)	26 (20.2%)	0.440

^aChi-square test/Fisher's exact test (p value < 0.05)

Fig. 1.

a Kaplan–Meier Estimator for time-to-cure by the variable CD20 neoplastic cells/10HPF (p = 0.0037). b Kaplan–Meier Estimator for time-to-death by the variable CD20 neoplastic cells/10HPF (p = 0.231)

Discussion

Predicting patients' prognosis is important in the clinical practice and is increasingly demanded in the anatomopathological report. The use of immunomarkers, such as CD20, their frequency and relationship to the clinical outcome have been the subject of research in recent decades.

The present study demonstrated that among CD20-positive patients, those with less than 10.5 neoplastic cells/10HPF achieve cure faster (p = 0.0037). Although controversial, most studies on CD20 immunoexpression have shown that this marker does not affect patients' prognosis [2–5, 10, 11]. Only two studies were associated with a good prognosis. Tzankov et al. [6] studied 119 patients with CHL followed for 12 years using a cutoff point positivity of 10% for CD20. According to these authors, the CD30 + /CD15 ± / CD20 + immunophenotype was associated with a favorable clinical outcome. They also demonstrated that CD20 is an independent positive prognostic factor (p = 0.035) and that disease-free survival was higher in CD20 + patients (286 months versus 202 months; p = 0.022). Canioni et al. [9], using a cutoff point positivity of 10% for CD20 in 200 randomly selected HRS cells, observed a lower frequency of CD20 positivity in refractory cases or those with early recurrence. However, in our study, we focused on a numerical and quantitative evaluation in absolute numbers, a more applicable method in the pathologist's practice than the percentage estimate used by other authors.

In conclusion, the present study demonstrated that in the group of CD20-positive patients, there is an association between CD20 positivity in neoplastic cells and a better clinical outcome with regard to disease-free survival, contrary to what is reported in most of the specialized literature. The present study, unlike previous studies, worked with a method of greater reproducibility in the healthcare practice.

Authors' Contributions

DFRL and CCO are responsible for the idea of the study, the respective plan of activities, review of clinical and pathological aspects of each biopsy. DFRL wrote the paper and CCO reviewed it. CCO oriented DFRL during the development of the research. All authors have approved the final article.

Funding

The authors declare that they have no financial or personal relationship(s) that may have inappropriately influenced them in writing this article.

Availability of Data and Materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

No conflict of interest to disclose.

Ethics Approval and Consent to Participate

The Clinical Hospital of Botucatu Medical School (HC-FMB-UNESP) research ethics committee approved the study under number 2.931.940.