|Ahead of print publication
Pattern of extranodal involvement and its impact on survival in diffuse large B-cell lymphoma from a tertiary cancer center in rural India
Chandran K Nair, Anju R Kurup, A Manuprasad, Praveen Kumar Shenoy, Vineetha Raghavan
Department of Clinical Hematology and Medical Oncology, Malabar Cancer Centre, Kannur, Kerala, India
|Date of Submission||19-Jun-2019|
|Date of Decision||13-Oct-2019|
|Date of Acceptance||01-Dec-2019|
|Date of Web Publication||19-Oct-2020|
Chandran K Nair,
Department of Clinical Hematology and Medical Oncology, Malabar Cancer Centre, Thalassery, Kannur - 670 103, Kerala
Source of Support: None, Conflict of Interest: None
Introduction: Extranodal (EN) involvement in diffuse large B-cell lymphoma (DLBCL) carries poor prognosis. Both the number and the specific sites of EN involvement are important in predicting prognosis. Given that the epidemiologic pattern of DLBCL in India is different from the rest of the world and such data correlations are scarce from developing countries, we aimed to find out if specific site and number of EN involvement could predict survival in DLBCL.
Methods: Patients with DLBCL treated with combination chemotherapy plus rituximab were included. Site and number of EN involvement were noted. Univariate analysis for survival was performed for EN involvement or not, specific site of involvement, and number of EN involvement (0/1 vs. ≥2).
Results: Among a total of 177 patients, 92 (52%) patients had EN disease. When patients with 2 or more EN sites were compared against patients with 0 or 1 site, there was significant reduction in both progression-free survival (PFS) and overall survival (OS) (3-year OS of 55% vs. 79%, P = 0.001, 3-year PFS of 42% vs. 65%, P = 0.001). When specific EN sites were studied for correlation with survival, involvement of skin/soft tissue, and serosa were associated with significant reduction in 3-year OS (33% vs. 74%, P = 0.011, and 63% vs. 75%, P = 0.03, respectively) and 3-year PFS (25% vs. 62%, P < 0.001, and 46% vs. 62%, P = 0.01, respectively).
Conclusion: Two or more EN sites in DLBCL predicted inferior survival. Serosal and skin/soft tissue involvement also predicted poor survival.
Keywords: Diffuse large B-cell lymphoma, extranodal, non-Hodgkin's lymphoma, pattern, survival
|How to cite this URL:|
Nair CK, Kurup AR, Manuprasad A, Shenoy PK, Raghavan V. Pattern of extranodal involvement and its impact on survival in diffuse large B-cell lymphoma from a tertiary cancer center in rural India. J Can Res Ther [Epub ahead of print] [cited 2021 Feb 25]. Available from: https://www.cancerjournal.net/preprintarticle.asp?id=298616
| > Introduction|| |
Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin's lymphoma (NHL) accounting for about 25% of NHL cases worldwide. Globally, NHL constitutes 2.8% of all cancer cases and amounts to 2.6% of all cancer-related mortality as per the Globocan 2018 data. The burden of NHL in India is different from the rest of the world with estimated incidence of about 2.2/100,000. The epidemiologic distribution in India is also a bit different from the western world with DLBCL constituting nearly 60% of all NHLs.,
Approximately 30%–40% of cases of DLBCL will have extranodal (EN) involvement at presentation. The presence of EN disease confers a poor prognosis in patients with DLBCL., It has also been shown that both the number and the specific sites of EN involvement are important in the prediction of prognosis in patients with DLBCL., However, such data correlations are scarce from developing countries. Moreover, as the epidemiologic pattern in India is different from the rest of the world, it will be worthwhile to study whether the same holds true in our population too. With this background, we aimed to study the correlation between number and sites of EN involvement with survival in patients with DLBCL.
| > Methods|| |
This is a retrospective analysis of all cases of newly diagnosed DLBCL, diagnosed and treated between January 2011 and June 2017. An official approval from the institutional review board was obtained before the start of the study. Inclusion criteria were set as (1) age ≥18 years, (2) newly diagnosed DLBCL, (3) treatment with rituximab plus CHOP or CEOP chemotherapy. Patients with primary mediastinal B-cell lymphoma, primary central nervous system (CNS) lymphoma, and primary cutaneous DLBCL-leg type were excluded. For patients with cardiac comorbidities, CEOP (Adriamycin replaced with etoposide) combined with rituximab was administered. Radiotherapy (RT) was administered in standard doses for limited-stage disease at the end of planned abbreviated courses of chemotherapy. Consolidation radiation was given to patients with advanced stages on a case to case basis depending on the presence of residual disease at the end of treatment. CNS prophylaxis was administered for cases with involvement of testis, vertebra, paranasal sinus, or other specific EN sites such as kidney/adrenals and bone marrow. CNS prophylaxis was with 4–6 doses of intrathecal methotrexate administered along with the chemotherapy schedules. Patients with testicular lymphoma diagnosed by orchiectomy specimen received RT to the opposite testis. Baseline characteristics of all patients were documented. EN involvement if any was noted. In patients with EN involvement, the exact numbers of EN involvement were noted.
As the policy of performing immunohistochemistry/molecular rearrangement studies for c-Myc, double hit or triple hit genetic changes and categorization into cell of origin subtypes  was introduced only recently in our institute, this aspect is not considered in this study. Response assessment was according to the Lugano criteria.
The normality of distribution of continuous variables was assessed with Shapiro–Wilk test. For finding out the differences in continuous variables between the two patient groups with and without extranodal involvement, either Student's t-test (if parametric distribution) or Mann Whitney U-test (if nonparametric distribution) were used. To compare differences in categorical variables, Fisher's exact test was used. For calculating survival duration, date of progression and date of death or date of the last follow-up of each patient were noted. The median follow-up was calculated using Kaplan–Meier (KM) curves by reversing event and censor codes. Overall survival (OS) was calculated from the date of diagnosis to date of death or date of the last follow-up. Progression-free survival (PFS) was calculated from the date of diagnosis to the date of progression or date of the last follow-up. Survival was analyzed by KM curves and Cox proportional hazards model. The analysis was performed with R v 3.2.2 (http://cran.r-project.org).
| > Results|| |
A total of 177 patients were identified as per study eligibility criteria. The median age was 58 (range 20–83). Majority of the patients were males with male-female ratio of 1.56. Baseline characteristics were shown in [Table 1]. Eighty-two (46%) patients had some form of comorbidities. One hundred and forty-three (81%) patients were in Eastern Cooperative Oncology Group Performance Status (ECOG-PS) <2. As shown in [Table 1], 112 patients (63%) were in advanced stage.
Ninety-two (52%) patients had EN disease. The most common site of EN involvement was bone/bone marrow in 43 patients (24%) followed by gastrointestinal tract (GIT) in 21 patients (12%) and serosa in 11 (6%). The involvement of sites such as parotid, breast, paranasal sinuses, female reproductive organs, and pancreas was in small numbers only and altogether constituted 14 patients (8%). The distribution of other EN involvement was as shown in [Table 2]. Patients with EN involvement, when compared against those without, had significantly higher advanced stages of disease, higher prevalence of bulky disease, and higher chance of progressing/becoming refractory to treatment. [Table 3] shows the comparison of baseline characteristics between patients with and without EN involvement.
The median follow-up was 22 months. Median OS and PFS were not reached for the entire cohort. Estimated 3-year OS was 75% (95% confidence interval [CI] 67–83). Estimated 3-year PFS was 61% (95% CI 52–70) [Figure 1]. There was no difference in OS between patients with and without EN involvement (3-year OS of 74% vs. 75% respectively, hazard ratio [HR] = 1.05, P = 0.8). There was a trend toward improved PFS in patients without EN involvement (3-year PFS of 57% vs. 63% in patients with and without EN involvement, respectively), even though it did not reach statistical significance (HR = 1.1, P = 0.7) [Figure 2].
|Figure 1: Overall survival and progression-free survival for the entire cohort|
Click here to view
|Figure 2: Overall survival and progression-free survival in patients with and without extranodal involvement|
Click here to view
The number of EN involvement was found to predict survival patterns. When patients with 2 or more EN sites were compared against patients with 0 or 1 EN involvement, there was significant reduction in both OS and PFS (3-year OS of 55% vs. 79%, HR = 3, P = 0.001, 3-year PFS of 42% vs. 65%, HR = 2.7, P = 0.001) [Figure 3]. When specific EN sites were studied for correlation with survival, involvement of skin/soft tissue (3-year OS of 33% vs. 75%, HR = 4.61, P = 0.011) and serosa (3-year OS of 62% vs. 75%, HR = 2.79, P = 0.03) predicted inferior OS [Figure 4]. Similarly, the involvement of skin/soft tissue (3-year PFS 0% vs. 62%, HR 8.2, P < 0.001) and serosa (3-year PFS of 46% vs. 62%, HR 2.84, P = 0.01) predicted inferior PFS also [Figure 5]. Significant difference in OS or PFS could not be demonstrated in patients with involvement of other EN sites such as testis, GIT, bone/bone marrow, liver, and kidney/adrenals.
|Figure 3: Overall survival and progression-free survival by number of extranodal sites|
Click here to view
|Figure 5: Progression-free survival by skin/soft tissue OR serosa involvement|
Click here to view
In univariate analysis, factors such as ECOG PS ≥2, Stage III/IV disease, high lactate dehydrogenase, advanced International Prognostic Index (IPI)/National Comprehensive Cancer Network (NCCN) IPI risk scores, and presence of B symptoms predicted inferior PFS. The same factors and age >60 predicted inferior OS [Table 4]. However, multivariate analysis could not replicate these findings.
| > Discussion|| |
The distribution of NHL subtypes is different in India and other Asian countries when compared to the western world. It has been found that DLBCL constitutes more than 60% of all NHLs in India, which is much higher than the reports from developed countries., The prevalence of follicular lymphoma and mantle cell lymphomas has been found to be much less common in India than in the United States and Europe. Considering these reasonably significant differences, we decided to see if the prognostic value of EN involvement in DLBCL holds true in our population too.
The presence of EN disease imparts poor treatment outcome in patients with DLBCL. This has been shown by various risk prognostication scoring systems such as IPI and NCCN IPI. Moreover, NCCN IPI has shown that more than just involvement, site-specific involvement also has a role to play in assessing the prognostic scores. However, all these data are from the western world. There is very little published data on this aspect from developing world including India. Hence, we undertook this study to know the effect of EN disease on treatment outcomes.
In our cohort, nearly half of the patients had EN involvement. This is much higher than the published literature form the western world. In the report by Møller et al., nearly 40% had EN involvement. Similarly, in the cohorts identified for NCCN IPI calculation and validation, the NCCN and BCCA cohorts had 36% and 25% EN involvement, respectively. As mentioned previously this difference could be part of the peculiar distribution pattern of NHL in India. Similar higher prevalence of EN involvement in DLBCL was also reported from other parts of Asia., One possible explanation for the difference in the distribution of EN involvement is the difference in tumor biologic and genetic factors across various countries and populations in the world.
We could demonstrate a trend toward lower PFS in patients with EN involvement, even though it did not reach statistical significance. The negative prognostic impact of EN involvement is well described. Hui et al. could demonstrate that the presence of any EN involvement (≥1 site) was associated with a reduction in both PFS and OS.
The original IPI risk score considers the presence of more than 1 EN site as one of the poor prognosis risk variables. We could also reach such a conclusion that the presence of 2 or more EN sites predicted inferior PFS and OS. A retrospective study from Singapore showed a similar pattern as ours with inferior survival in patients with >1 EN involvement in univariate analysis. Yet another study from Korea described that ≥3 EN sites were a better threshold than ≥2. All these Asian studies included patients who had received rituximab-containing regimens. However, there were many published reports from the west which could not replicate this IPI pattern in the rituximab-treated patients., The disparity between the reports from Asian studies, including our present one and the western world could be explained by (1) the different prevalences of EN involvement and (2) the variation in distribution of NHLs among different geographical regions and ethnicity.
For predicting prognosis in patients with DLBCL in the rituximab era, Zhou et al. formulated and validated NCCN IPI. In this prognostic index, prognostic significance of the specific site of EN involvement was demonstrated. Later, a few more publications came out seeking for similar correlation.,, The most common site of EN involvement in our study was bone/bone marrow. All the above-mentioned studies in this context had different patterns of EN involvement. Moreover, the types of EN involvement predicting poor outcome was also different. In our study, the involvement of serosa or skin/soft tissue predicted inferior PFS and OS. The different patterns in involvement and prognostic values can again be explained by the geographical variation.
| > Conclusion|| |
We could show that more than 1 site of EN involvement predicted poor survival in DLBCL, as in the original IPI risk stratification, even in rituximab-based treatment groups. The involvement of skin/soft tissue or serosa also predicted inferior survival. This study has definite limitations in the form of being retrospective and based on a single institution data and having relatively small sample size. However, we strongly believe that our data could throw some light on the prognostic aspect of EN involvement in DLBCL from a developing country like India.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424.
Nair R, Arora N, Mallath MK. Epidemiology of non-Hodgkin's lymphoma in India. Oncology 2016;91 Suppl 1:18-25.
Gogia A, Das CK, Kumar L, Sharma A, Sharma MC, Mallick S. Profile of non-Hodgkin lymphoma: An Indian perspective. South Asian J Cancer 2018;7:162.
] [Full text]
Møller MB, Pedersen NT, Christensen BE. Diffuse large B-cell lymphoma: Clinical implications of extranodal versus nodal presentation – A population-based study of 1575 cases. Br J Haematol 2004;124:151-9.
Hui D, Proctor B, Donaldson J, Shenkier T, Hoskins P, Klasa R, et al
. Prognostic implications of extranodal involvement in patients with diffuse large B-cell lymphoma treated with rituximab and cyclophosphamide, doxorubicin, vincristine, and prednisone. Leuk Lymphoma 2010;51:1658-67.
International Non-Hodgkin's Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin's lymphoma. N
Engl J Med 1993;329:987-94.
El-Galaly TC, Villa D, Alzahrani M, Hansen JW, Sehn LH, Wilson D, et al
. Outcome prediction by extranodal involvement, IPI, R-IPI, and NCCN-IPI in the PET/CT and rituximab era: A Danish-Canadian study of 443 patients with diffuse-large B-cell lymphoma. Am J Hematol 2015;90:1041-6.
Zhou Z, Sehn LH, Rademaker AW, Gordon LI, Lacasce AS, Crosby-Thompson A, et al
. An enhanced International Prognostic Index (NCCN-IPI) for patients with diffuse large B-cell lymphoma treated in the rituximab era. Blood 2014;123:837-42.
Damlaj M, Yassin R, Al-Shieban S, Pasha T, Alzahrani M, Alhejazi A, et al
. BCL-2 overexpression overcomes cell of origin stratification in diffuse large B-cell lymphoma. Hematol Oncol Stem Cell Ther 2019;12:174-7.
Cheson BD, Fisher RI, Barrington SF, Cavalli F, Schwartz LH, Zucca E, et al
. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: The Lugano classification. J Clin Oncol 2014;32:3059-68.
Naresh KN, Srinivas V, Soman CS. Distribution of various subtypes of non-Hodgkin's lymphoma in India: A study of 2773 lymphomas using R.E.A.L. and WHO Classifications. Ann Oncol 2000;11 Suppl 1:63-7.
Yoo C, Kim S, Sohn BS, Kim JE, Yoon DH, Huh J, et al
. Modified number of extranodal involved sites as a prognosticator in R-CHOP-treated patients with disseminated diffuse large B-cell lymphoma. Korean J Intern Med 2010;25:301-8.
Yao S, Li J, Yao Z, Xu Y, Chu J, Zhang J, et al
. Extranodal involvement in young patients with diffuse large B-cell lymphoma: Distribution, prognostic value and treatment options. Chin J Cancer Res 2017;29:57-65.
Wallace TA, Martin DN, Ambs S. Interactions among genes, tumor biology and the environment in cancer health disparities: Examining the evidence on a national and global scale. Carcinogenesis 2011;32:1107-21.
Ngo L, Hee SW, Lim LC, Tao M, Quek R, Yap SP, et al
. Prognostic factors in patients with diffuse large B cell lymphoma: Before and after the introduction of rituximab. Leuk Lymphoma 2008;49:462-9.
Ziepert M, Hasenclever D, Kuhnt E, Glass B, Schmitz N, Pfreundschuh M, et al
. Standard International prognostic index remains a valid predictor of outcome for patients with aggressive CD20+ B-cell lymphoma in the rituximab era. J Clin Oncol 2010;28:2373-80.
Lu CS, Chen JH, Huang TC, Wu YY, Chang PY, Dai MS, et al
. Diffuse large B-cell lymphoma: Sites of extranodal involvement are a stronger prognostic indicator than number of extranodal sites in the rituximab era. Leuk Lymphoma 2015;56:2047-55.
Hwang HS, Yoon DH, Suh C, Huh J. A new extranodal scoring system based on the prognostically relevant extranodal sites in diffuse large B-cell lymphoma, not otherwise specified treated with chemoimmunotherapy. Ann Hematol 2016;95:1249-58.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3], [Table 4]