Indian Journal of Pathology and Microbiology
Home About us Instructions Submission Subscribe Advertise Contact e-Alerts Ahead Of Print Login 
Users Online: 239
Print this page  Email this page Bookmark this page Small font sizeDefault font sizeIncrease font size


 
  Table of Contents    
BRIEF COMMUNICATION  
Year : 2017  |  Volume : 60  |  Issue : 1  |  Page : 87-91
Identification of prognostic factors in patients with diffuse large B-cell lymphoma


Department of Pathology, The First Hospital of Jilin University, Changchun, China

Click here for correspondence address and email

Date of Web Publication14-Feb-2017
 

   Abstract 

To identify prognostic factors for patients with diffuse large B-cell lymphoma (DLBCL), specifically those classified into conflicting subgroups by Hans' and Choi's classification algorithms. We retrospectively reviewed clinical and pathological data of 154 patients diagnosed with de novo DLBCL in the First Hospital of Jilin University from January 2004 to September 2011. All cases were classified into subgroups based on Hans' and Choi's algorithms with immunohistochemical markers. Statistical Analysis Used: The correlation between various clinicopathological factors and 5-year survival rate, the correlation between those factors with the International Prognostic Index, the concordance between Hans' and Choi's approach was evaluated. The survival in different subtypes as classified by Hans' or Choi's approach was mapped. Results: The Eastern Cooperative Oncology Group (ECOG) performance score 2–5, positive Bcl-2 expression, negative CD10 expression or negative Bcl-6 expression significantly correlated with worse prognosis. The two algorithms showed good consistency (83% concordance, Kappa = 0.660, P < 0.001). By both classifications, the 5-year overall survival rate in germinal center B-cell-like subtype (GCB) lymphoma is significantly higher than that in the non-GCB subtype. There were 25 cases assigned to conflicting subtypes by the two approaches. Among these 25 cases, ECOG 2–5, positive Bcl-2 expression, negative CD10 expression, or negative Bcl-6 expression significantly correlated with worse prognosis. Conclusions: ECOG 2–5, positive Bcl-2 expression, negative CD10 expression, or negative Bcl-6 expression are independent markers for poor prognosis of DLBCL patients. There were 15% cases assigned to conflicting subgroups based on the two algorithms. For these cases, ECOG 2–5, positive Bcl-2 expression, negative CD10 expression, or negative Bcl-6 expression still significantly correlate with poor prognosis.

Keywords: Diffuse large B-cell lymphoma, germinal center B-cell-like subtype, immunophenotyping, international prognostic index, overall survival, prognosis

How to cite this article:
Peng F, Guo L, Yao WK, Zheng Y, Liu Y, Duan XM, Wang YP. Identification of prognostic factors in patients with diffuse large B-cell lymphoma. Indian J Pathol Microbiol 2017;60:87-91

How to cite this URL:
Peng F, Guo L, Yao WK, Zheng Y, Liu Y, Duan XM, Wang YP. Identification of prognostic factors in patients with diffuse large B-cell lymphoma. Indian J Pathol Microbiol [serial online] 2017 [cited 2017 May 27];60:87-91. Available from: http://www.ijpmonline.org/text.asp?2017/60/1/87/200056



   Introduction Top


Diffuse large B-cell lymphoma (DLBCL) is the most common form of B-cell-derived lymphoma, constituting 30%–40% of non-Hodgkin lymphoma in adults.[1] DLBCL presents significant heterogeneity with respect to the underlying genetics, histological morphologies, immunophenotypes, clinical manifestations, and thus there is variation in responses to treatment as well as prognosis.[2] Therefore, it is important to classify DLBCL according to distinct clinicopathological parameters and prognostic markers, to develop efficient therapies tailored to each DLBCL subtype. In clinical practice, the prognosis of DLBCL patients is often estimated using the International Prognostic Index (IPI),[3] which reflects a mixture of underlying biologic or genetic differences. However, it has been shown that even with the same IPI score, DLBCL patients experience distinct clinical courses. To address this issue, two general classification algorithms have been developed by Hans et al.[4] and Choi et al.,[5] respectively. Both algorithms divide DLBCL cases into germinal center B-cell-like (GCB) subtype with good prognosis and non-GCB subtype with poor prognosis based on immunohistochemical analysis. Hans' approach focused on the expressions of CD10, Bcl-6, and MUM-1 [Figure 1] while Choi's approach added two more targets, GCET1, and FOXP1 [Figure 2].
Figure 1: The rationale of Hans' classification to separate diffuse large B-cell lymphoma patients into germinal center B-cell-like subtype with good prognosis and non-germinal center B-cell-like subtype with poor prognosis by immunophenotyping of tumor cell markers CD10, Bcl-6, and MUM-1

Click here to view
Figure 2: The rationale of Choi's classification with the addition of two tumor cells markers, GCET1, and FOXP1, on those used by Hans' classification

Click here to view


Although both Hans' and Choi's algorithms have worked reasonably well, they do yield conflicting results, that is, a case classified by Hans' algorithm as GCB subtype with good prognosis becomes a non-GCB subtype with poor prognosis by Choi's method and vice versa. Such conflicts may lead to the selection of wrong treatments for the patients. In this study, we retrospectively analyzed 154 DLBCL patients and identified prognosis-predicting factors for DLBCL, specifically for those showing conflicting results when classified by Hans' versus Choi's classification algorithms.


   Subjects and Methods Top


Patients

We studied 154 patients with de novo DLBCL and complete clinicopathological and follow-up information in the First Hospital of Jilin University (Changchun, China) from January 2004 to September 2011. All diagnoses of DLBCL conformed to the 2008 WHO criteria. This study was approved by the Ethics Committee of Jilin University, and written consent was obtained from all participating patients.

Reagents

The primary antibodies were purchased from Maxvision (Fuzhou, China) and ZSGB-Bio (Beijing, China).

Immunohistochemistry

We performed all immunohistochemical staining on 4-µm-thick, 10% neutral buffer formalin-fixed, paraffin-embedded specimens using the streptavidin-peroxidase method.

International Prognostic Index scoring

Each patient was given an IPI score ranging from 0 to 5, with one point assigned to each of the following risk factors: Age over 60, clinical stage III/IV ≥2 extranodal lesions, the Eastern Cooperative Oncology Group (ECOG) score ≥2, and elevated serum lactate dehydrogenase.

Statistical analysis

The Kaplan–Meier method with the log-rank test was used to map and compare the relationships between the overall survival and the cumulative survival rate. The correlation between the clinicopathological parameters and IPI was assessed using the paired Chi-square test. The consistency between the Hans' and Choi's classification algorithms was analyzed by the Kappa test. Multivariate analysis was performed using the Cox regression method. All statistical analyses were performed with SPSS 18.0 software (IBM SPSS Inc, Chicago, USA), with a P ≤ 0.05 being considered statistically significant.


   Results Top


Identification of individual clinicopathological factors correlated to 5-year overall survival

The analysis demonstrated that the 5-year overall survival rate was significantly lower in patients with ECOG 2–5 than in those with ECOG 0–1 (P = 0.009), in patients with positive Bcl-2 (P = 0.001), MUM1 (P = 0.001), or FOXP1 (P < 0.001) expression than in those with negative expression and in patients with negative CD10 (P < 0.001) or Bcl-6 (P = 0.022) expression than in those with positive expression [Table 1]. In contrast, other clinicopathological factors showed no significant correlation with patient survival (P > 0.05) [Table 1].
Table 1: The correlation between individual clinicopathological factors and prognosis in 154 diffuse large B-cell lymphoma cases

Click here to view


The analysis showed that patients with ECOG 2–5 had significantly worse prognosis and a 1.756-fold higher risk of death than those with ECOG 0–1 (P = 0.021). In addition, positive Bcl-2 expression, negative CD10, and Bcl-6 expression all significantly correlated with worse prognosis (risk ratio = 2.494, 2.237, and 1.698, respectively; P = 0.008, 0.003, and 0.046, respectively) [Table 2].
Table 2: Cox regression analysis between multiple clinicopathological factors and prognosis

Click here to view


Identification of individual clinicopathological factors correlated to International Prognostic Index

We found that higher ECOG score (2–5, P < 0.001), positive Bcl-2 expression (P = 0.003), negative CD10 (P = 0.035), or Bcl-6 (P < 0.001) all significantly correlated with a higher IPI score (3–5) [Table 3].
Table 3: The relevance between clinicopathological factors and international prognostic index in 154 diffuse large B.cell lymphoma patients

Click here to view


Assessment of the consistency between Hans' and Choi's classification algorithms

There were 66 GCB (42.9%) and 88 non-GCB (57.1%) cases according to Hans' classification. Choi's approach divided the 154 DLBCL patients into GCB (53 patients, 34.4%) and non-GCB (101 patients, 65.6%) group. When these two algorithms were compared, 47 cases (30.5%) were classified as GCB and 82 cases (53.2%) as non-GCB by both approaches, whereas 25 cases (16.2%) were classified into conflicting groups from one approach versus the other [Table 4]. Overall, these two approaches showed a good consistency at 83.8% (Kappa = 0.660, P < 0.001) [Table 4]. For both approaches, the 5-year overall survival of GCB patients was significantly higher than the non-GCB types (75% vs. 43%, P < 0.001 for Hans' approach; 83% vs. 44%, P < 0.001 for Choi's approach) [Figure 3].
Table 4: The consistency test of Hans' and Choi's classification algorithms

Click here to view
Figure 3: By both Hans' (a) and Choi's (b) classification, patients with germinal center B-cell-like subtype present significantly better prognosis than those with non-germinal center B-cell-like subtype subtype

Click here to view


Identification of clinicopathological factors correlated with International Prognostic Index in cases assigned to conflicting subgroups by Hans' versus Choi's classification algorithms

Among the 25 patients assigned to conflicting subgroups, 13 were male and 12 female, their average age was 56 years old, ranging from 9 to 72 years. During the 5-year follow-up, 11 (44%) patients passed away.

The analysis showed that those with ECOG 2–5 had a higher proportion in the high-risk group (IPI 3–5) compared with those with ECOG 0–1 (76.9% vs. 23.1%, P < 0.001). Similarly, positive Bcl-2 expression (92.3% vs. 7.7%, P = 0.041), negative CD10 expression (84.6% vs. 15.4%, P = 0.015), and negative Bcl-6 expression (84.6% vs. 15.4%, P = 0.015) were all seen with a higher frequency than their counterparts in the high-risk group [Table 5].
Table 5: The clinicopathological characters of cases with opposite results and their relevance to the international prognostic index

Click here to view



   Discussion Top


In the current clinical practice, the prognosis of DLBCL patients is assessed using clinical parameters of IPI. However, the conflicts in patient prognosis with the same IPI have prompted to develop novel classification algorithms.[6] Among these algorithms, Hans' and Choi's method is commonly accepted.

In this study, we found that by both univariate and multivariate analyses, ECOG 2–5, positive Bcl-2 expression, negative CD10 expression, and negative Bcl-6 expression were independent markers for poor prognosis. When applying the Hans' and Choi's algorithms to these patients, 25 were classified into conflicting subgroups by these two methods. Further analysis on these 25 patients revealed that ECOG 2–5, positive Bcl-2 expression, negative CD10 expression, or negative Bcl-6 expression still function effectively as markers for poor prognosis.

In our investigation, ECOG 2–5 was independently associated with worse prognosis and higher IPI score in 154 DLBCL patients as well as the 25 patients classified into conflicting subgroups by Hans' and Choi's algorithms. Consistently, Wang et al. showed that ECOG 2–5 was an independent prognostic factor for the worse prognosis of patients with primary testicular DLBCL.[7] Markovic et al. demonstrated that ECOG correlated with the therapeutic responses and prognosis of DLBCL patients.[8] One potential explanation for the association between ECOG and DLBCL prognosis is that patients with low ECOG have good physical capacity, allowing them to better tolerate the side effects of chemotherapy compared with those with a high ECOG score, enabling long-term chemotherapy, and subsequently longer survival.

Consistent with the previous studies, we showed that positive Bcl-2 expression is an independent prognostic marker for significantly lower 5-year survival. The death risk in patients with positive Bcl-2 expression was 2.5 times higher than those negative for Bcl-2, similar to the finding by Biasoli et al. (risk ratio = 2.43).[9] Given that patients with Bcl-2-positive expression were resistant to CHOP chemotherapy combining rituximab, a negative regulator for Bcl-2, with CHOP overcame Bcl-2 resistance, and significantly improved the treatment efficacy on a subgroup of elderly patients with DLBCL.[10] Therefore, detecting the expression of Bcl-2 might provide critical information on treatment selection for DLBCL patients. In our hospital, we performed Bcl-2 expression analysis by immunohistochemistry as a routine test for DLBCL patients.

Among 25 DLBCL cases classified into conflicting subgroups, 16 cases showed positive expression of CD10 (64%). In addition, the negative expression of CD10 significantly correlated with a higher IPI score (IPI 3–5). Similarly, negative CD10 expression significantly correlated with lower 5-year overall survival and higher IPI, demonstrating CD10 as an independent prognostic factor DLBCL. For the underlying mechanisms, a previous study showed that CD10 promotes apoptosis by degrading cytokines induced by antiapoptotic signals,[11] which is corroborated by a positive correlation between CD10 expression and apoptosis of B lymphoma cells.[11]

The present study further supported the notion that Bcl-6 is an independent factor for prognosis, which might be achieved by inhibiting cell differentiation and inducing cell activation as well as proliferation. In addition, aberrant Bcl-6 expression prolonged the cell cycle and increased cell instability leading to malignant transformation.[12]


   Conclusions Top


We investigated the correlations between various clinicopathological factors and prognosis of 154 DLBCL patients as well as the 25 DLBCL patients that were classified into conflicting subgroups according to Hans' and Choi's immunophenotyping approach. We found that ECOG 2–5, positive expression of Bcl-2, and negative expression of CD10 or Bcl-6 are independent clinicopathological factors for poor prognosis. These factors provide valuable insights for predicting prognosis and directing treatment selection, and therefore, will benefit the survival of patients with DLBCL.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

 
   References Top

1.
Sabattini E, Bacci F, Sagramoso C, Pileri SA. WHO classification of tumours of haematopoietic and lymphoid tissues in 2008: An overview. Pathologica 2010;102:83-7.  Back to cited text no. 1
    
2.
Gascoyne RD. Molecular heterogeneity of diffuse large B-cell lymphoma. Hematol J 2004;5 Suppl 3:S144-8.  Back to cited text no. 2
    
3.
A predictive model for aggressive non-Hodgkin's lymphoma. The International Non-Hodgkin's Lymphoma Prognostic Factors Project. N Engl J Med 1993;329:987-94.  Back to cited text no. 3
    
4.
Hans CP, Weisenburger DD, Greiner TC, Gascoyne RD, Delabie J, Ott G, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood 2004;103:275-82.  Back to cited text no. 4
    
5.
Choi WW, Weisenburger DD, Greiner TC, Piris MA, Banham AH, Delabie J, et al. A new immunostain algorithm classifies diffuse large B-cell lymphoma into molecular subtypes with high accuracy. Clin Cancer Res 2009;15:5494-502.  Back to cited text no. 5
    
6.
Shipp MA, Ross KN, Tamayo P, Weng AP, Kutok JL, Aguiar RC, et al. Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nat Med 2002;8:68-74.  Back to cited text no. 6
    
7.
Wang Y, Li ZM, Huang JJ, Xia Y, Li H, Li YJ, et al. Three prognostic factors influence clinical outcomes of primary testicular lymphoma. Tumour Biol 2013;34:55-63.  Back to cited text no. 7
    
8.
Markovic O, Marisavljevic D, Cemerikic V, Perunicic M, Savic S, Filipovic B, et al. Clinical and prognostic significance of apoptotic profile in patients with newly diagnosed nodal diffuse large B-cell lymphoma (DLBCL). Eur J Haematol 2011;86:246-55.  Back to cited text no. 8
    
9.
Biasoli I, Morais JC, Scheliga A, Milito CB, Romano S, Land M, et al. CD10 and Bcl-2 expression combined with the International Prognostic Index can identify subgroups of patients with diffuse large-cell lymphoma with very good or very poor prognoses. Histopathology 2005;46:328-33.  Back to cited text no. 9
    
10.
Mounier N, Briere J, Gisselbrecht C, Emile JF, Lederlin P, Sebban C, et al. Rituximab plus CHOP (R-CHOP) overcomes bcl-2 – Associated resistance to chemotherapy in elderly patients with diffuse large B-cell lymphoma (DLBCL). Blood 2003;101:4279-84.  Back to cited text no. 10
    
11.
Ye ZY, Cao YB, Lin TY, Lin HL. Subgrouping and outcome prediction of diffuse large B-cell lymphoma by immunohistochemistry. Zhonghua Bing Li Xue Za Zhi 2007;36:654-9.  Back to cited text no. 11
    
12.
Basso K, Dalla-Favera R. Roles of BCL6 in normal and transformed germinal center B cells. Immunol Rev 2012;247:172-83.  Back to cited text no. 12
    

Top
Correspondence Address:
Yin-Ping Wang
Department of Pathology, The First Hospital of Jilin University, 71 Xinmin Street, Changchun, Jilin Province
China
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/0377-4929.200056

Rights and Permissions


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5]



 

Top
 
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  


    Abstract
   Introduction
   Subjects and Methods
   Results
   Discussion
   Conclusions
    References
    Article Figures
    Article Tables

 Article Access Statistics
    Viewed709    
    Printed10    
    Emailed0    
    PDF Downloaded34    
    Comments [Add]    

Recommend this journal