Keywords
Burkitt's lymphoma - non-Hodgkin lymphoma - clinical characteristics - treatment outcome
Introduction
Burkitt's lymphoma (BL) is an aggressive B-cell lymphoma characterized by a high degree
of proliferation of the malignant cells and deregulation of the proto-oncogene c-myc.[1] BL is predominantly a disease of childhood, but it is also seen in the adult population.
The World Health Organization (WHO) classification of BL describes three clinical
variants: endemic, sporadic, and immunodeficiency-related.[2] These types are similar in morphology, immunophenotype, and genetic features. Endemic
BL is the most common childhood cancer in equatorial Africa associated with Epstein–Barr
virus (EBV) and plasmodium falciparum malaria. The sporadic type predominates in the
rest of the world with no special climatic or geographical links, and is rarely associated
with EBV infection. Sporadic BL (sBL), accounts for 1 to 2% of all adult lymphomas
in Western Europe and the United States.[3] The diagnostic and therapeutic principles of BL in adults are the same as those
in the pediatric population. A recent study of the Swedish Lymphoma Registry confirmed
advanced age, poor WHO performance status, and elevated serum lactate dehydrogenase
(LDH) as significant prognostic factors for adult BL.[4]
The present retrospective analysis provides information on clinical presentation,
histology patterns, and outcome of adolescents, adults, and elderly patients with
BL treated at a tertiary cancer center in India.
Objectives
The objectives of this article were to study the clinical characteristics, treatment
response, and survival of BL in adolescent and adult patients. The primary outcome
measures were treatment response and progression-free survival (PFS). The secondary
outcome measure was overall survival (OS) at 8 years.
Materials and Methods
Study Setting: This was a retrospective study conducted in the department of medical oncology at
a tertiary cancer center in India during a 10-year period (January 2005 to December
2014). Patients with BL/leukemia above 14 years of age treated during the study period
were included. We analyzed 60 patients with BL.
Inclusion/ Exclusion Criteria: Eligibility criteria included all patients aged above 14 years with a histological
diagnosis of BL/leukemia. Patients with relapsed BL and who received some prior treatments
were excluded from the study.
Methodology: Medical records of patients were studied with respect to the demographic details,
clinical history, physical examination, baseline investigations like complete hemogram,
serum chemistry, and serum LDH. Staging workup including histopathology report, bone
marrow study, cerebrospinal fluid analysis, and imaging studies were noted. The disease
was staged according to the Ann Arbor staging system. The treatment response at the
end of induction and on completion of treatment was obtained. Clinical response was
classified as complete remission (CR), partial remission (PR), stable disease (SD),
and progressive disease based on modified Cheson lymphoma response evaluation criteria.[5] OS was assessed from the initiation of definitive chemotherapy to the last follow-up
or death and PFS was calculated from the initiation of chemotherapy till disease progression.
Statistical Analysis: The baseline patient characteristics, treatment details, and response assessment
were analyzed using descriptive statistics. OS and PFS were obtained by Kaplan–Meier
method, using SPSS v. 11. The risk analysis for OS and PFS was done using Cox regression
analysis. Statistical significance was defined as a p-value less than 0.05.
Ethics: The study was approved by the Institutional Review Board (IRB No. 09/2016/03, dated
September 1, 2016). Informed consent was waived off due to retrospective nature of
the study. All procedures performed in study were in accordance with the ethical standards
of the institutional and/or national research committee and with the 1964 Helsinki
Declaration and its later amendments or comparable ethical standards.
Results
Baseline Patient Characteristics (shown in [Table 1])
Table 1
Baseline patient characteristics
|
Baseline patient characteristics (n = 60)
|
Frequency, n (%)
|
|
Median age (years)
|
42 (range: 14–81)
|
|
M: F ratio
|
2.16: 1
|
|
B symptoms
|
39 (65)
|
|
Hb < 10 g/ dL
|
15 (25)
|
|
Platelet count
< 1,00,000/mm3
|
06 (10)
|
|
Elevated LDH
|
44 (73)
|
|
Biochemical TLS
(S.UA≥8mg/dL, S. K≥6mEq/L, S. Ph≥4.6mg/dL, S. Ca ≤ 7mg/dL)
|
13 (22)
|
|
Hypoalbuminemia
(S. albumin < 3.5 g/dL)
|
13 (22)
|
|
Renal impairment (S. Cr > 1.5 of ULN)
|
06 (10)
|
|
Hepatic impairment (S. Bil > 2 times of ULN)
|
7 (12)
|
|
Stage
I
II
III
IV
|
17 (28)
16 (27)
05 (8)
22(37)
|
|
Burkitt's leukemia (bone marrow blast > 20%)
|
11(18)
|
Abbreviations: Hb, hemoglobin; LDH, lactate dehydrogenase; M: F, male: female ratio;
S. Bil, serum bilirubin; S. Ca, serum calcium; S. Cr, serum creatinine; S. K, serum
potassium; S. Ph, serum phosphorus; S.UA, serum uric acid; TLS, tumor lysis syndrome;
ULN, upper limit of normal.
During the period 2005 to 2014, 3,811 cases of non-Hodgkin lymphoma (NHL) above 14
years of age were treated in our department, of which 60 were BL (1.57%). The median
age at presentation was 42 years (range: 14–81 years). There were 41 males and 19
females, with the male to female ratio of 2.16: 1. Twenty-five patients (42%) were
less than 40 years of age and elderly patients above more than 65 years constituted
10% of our study population. The most common presenting symptoms were lymphadenopathy,
abdominal pain and abdominal distension. Two patients each presented with paraparesis,
breast lump, jaw swelling and one patient presented with vaginal mass. ([Supplementary Material Table A], available in the online version) Three patients presented to us following surgery
for acute intestinal obstruction and diagnosis of BL was confirmed on histopathological
examination of the surgically resected specimen. B symptoms were present in 39 patients
(65%). The median duration of symptoms was 4 weeks (range: 1–16 weeks). Eleven patients
had BL at presentation (18%) and 13 patients had features of tumor lysis syndrome
(22%). Laboratory tumor lysis syndrome was present in all 13 patients and clinical
tumor lysis syndrome in 6 patients (10%). Patient had a mean hemoglobin of 11.6 g/dL,
mean total leucocyte count of 8,586/mm3, and a mean platelet count of 2.8 lakhs/mm3. In our study population, only one patient tested positive for human immunodeficiency
virus (HIV) infection. Cerebrospinal fluid was positive for malignant cells in two
patients (3%). Twenty-nine patients had extra nodal disease with bone marrow being
the most common site (48%). The Ann Arbor stage was I in 17 (28%), II in 16 (27%),
III in 5 (8%), and IV in 22 (37%).
All patients had a histopathological diagnosis either from lymph node biopsy or biopsy
from the involved organ. Morphologically the tumor cells were seen in sheets having
scanty cytoplasm with clumped chromatin, surrounded by normal phagocytic cells having
“starry sky pattern.” Immunohistochemically (IHC) the tumor cells were positive for
CD 10, CD 19, CD20, BCL6 and S. immunoglobulin M and negative for terminal deoxynucleotidyl
transferase. Complete IHC marker panel was done in 40 patients. Out of 11 patients
with BL, four patients were diagnosed as acute lymphoblastic leukemia L3 on flow cytometry
([Supplementary Material Table B], available in the online version). Thirty-nine patients had a MIB labeling index
of 100% and the lowest MIB labeling index was 80%.
Treatment Characteristics ([Table 2])
Table 2
Treatment summary
|
Treatment outcome
|
HCVAD ± R
|
CHOP/CVP ± R
|
BFM
|
R CODOX
M-IVAC
|
Murphy's protocol
|
|
Total no.
|
35
|
13
|
4
|
1
|
2
|
|
Complete remission
|
22 (63%)
|
6 (47%)
|
3 (75%)
|
1 (100%)
|
2 (100%)
|
|
Progressive disease
|
9 (26%)
|
3 (23%)
|
1 (25%)
|
Nil
|
Nil
|
|
Death
|
4 (11%)
|
4 (30%)
|
Nil
|
Nil
|
Nil
|
|
Relapse
|
2 (6%)
|
NIL
|
1 (25%)
|
Nil
|
1 (50%)
|
|
8-year OS
|
60%
|
46.2%
|
50%
|
100%
|
50%
|
Abbreviations: BFM; standard Berlin-Frankfurt-Munich protocol; CHOP/ CVP ± R; cyclophosphamide,
adriamycin, vincristine, prednisolone/ cyclophosphamide, vincristine, prednisolone ± rituximab;
HCVAD ± R; cyclophosphamide, vincristine, adriamycin, dexamethasone ± rituximab; OS,
overall survival; R CODOX M- IVAC, rituximab ± cyclophosphamide, doxorubicin, vincristine,
methotrexate, etoposide, ifosfamide and cytarabine.
Among the 60 patients, 55 (92%) received treatment and five refused to get treated.
All received combination chemotherapy that included hyper-CVAD ± rituximab (HCVAD ± R)
in 35 (63%), CHOP/ CVP ± rituximab (CHOP/CVP ± R) in 13 (24%), Berlin-Frankfurt-Munich
protocol (BFM) in 4 (7%), Murphy's protocol in 2(4%), and rituximab ± cyclophosphamide,
doxorubicin, vincristine, methotrexate, etoposide, ifosfamide and cytarabine (R CODOX
M-IVAC) in 1 (2%) patient depending on the time period when the diagnosis was made.
Patients with poor performance status (ECOG - Eastern Co-operative Oncology Group
performance status 3 or 4) and advanced age (age > 65 years) received CHOP/ CVP ± R
(median age 69) and young and fit patients received intensive chemotherapy (median
age 32). Thirty-four patients attained CR, 33 after chemotherapy, and 1 patient following
radiation (RT) after chemotherapy. Thirteen patients had progressive disease and eight
patients died during chemotherapy. Among the 34 (62%) patients who attained CR, 22
received HCVAD ± R, 6 had CHOP/ CVP ± R, 3 had BFM, 1 had R CODOX M-IVAC, and 2 patients
received Murphy's protocol.
Thirty-five patients received HCVAD ± R among them, 22 (63%) attained complete remission,
9 (26%) had progressive disease, 2 (6%) relapsed, and 4 (11%) patients died during
the treatment. The median cycles of HCVAD regimen were 8 (4 cycles of A and 4 cycles
of B). Twenty-two out of thirty-five patients (63%) were able to complete all planned
cycles of chemotherapy with a median duration of 7 months (range: 6–10 months). Out
of thirteen patients who received CHOP/ CVP ± R chemotherapy, only six (46%) patients
attained complete remission, three (23%) had disease progression, and four patients
(31%) died during chemotherapy. All patients on R CODOX-M/IVAC, Murphy's protocol
and three of four patients on BFM protocol attained complete remission.
Among the 34 patients who achieved remission, four patients (12%) relapsed and received
palliative chemotherapy later. Two patients with central nervous system (CNS) disease
at diagnosis, two with extra nodal primary and one patient with residual disease after
chemotherapy received RT after completion of the chemotherapy.
Table 3
Survival data
|
Variables
|
8-year PFS (%)
|
8-year OS (%)
|
|
%
|
SE
|
p-Value
|
%
|
SE %
|
p-Value
|
|
Age (years)
< 40
40–65
> 65
|
80.0
41.2
16.7
|
8.0
10.1
15.2
|
0.001
|
80.0
45.8
16.7
|
8.0
10.2
15.2
|
0.001
|
|
Sex
Male
Female
|
52.5
66.0
|
7.9
12.4
|
0.301
|
55.0
66.7
|
7.9
12.2
|
0.299
|
|
LDH
< 2000
≥ 2000
|
58.3
69.2
|
8.2
12.8
|
0.573
|
61.1
69.2
|
8.1
12.8
|
0.558
|
|
Burkitt's lymphoma
Burkitt's leukemia
|
56.8
54.5
|
7.5
15.0
|
0.885
|
59.1
54.5
|
7.4
15.0
|
0.87
|
|
Stage
1
2
3
4
|
56.3
42.9
100
58.7
|
12.4
13.2
-
10.6
|
0.425
|
62.5
42.9
100
59.1
|
12.1
13.2
-
10.5
|
0.430
|
|
Chemo
CHOP ± R
HCVAD ± R
|
36.9
60.0
|
13.8
8.3
|
0.123
|
46.2
60.0
|
13.8
8.3
|
0.125
|
Abbreviations: CHOP/ CVP ± R; cyclophosphamide, adriamycin, vincristine, prednisolone/
cyclophosphamide, vincristine, prednisolone ± rituximab; HCVAD ± R; cyclophosphamide,
vincristine, adriamycin, dexamethasone ± rituximab; LDH, lactate dehydrogenase; OS,
overall survival; PFS, progression-free survival; SE, standard error.
The PFS was 58.3 at 2 years and 56.2 at 5 and 8 years. The 2-year OS was 61.6% and
58.2% at 5 and 8 years ([Fig. 1]). The 8-year OS for patients with age group less than 40 years, 40 to 65 years,
and more than 65 years were 80, 41.2, and 16.7%, respectively. Both OS and PFS were
significantly high in adolescents and young adults compared to patients above 40 years
of age (p-value of 0.001). There was no significant survival difference with respect to sex,
stage of disease, serum LDH, and site of involvement. Patients who received HCVAD-based
chemotherapy had better survival compared with CHOP/CVP chemotherapy even though it
was not statistically significant (8 years OS of 60.0% ± 8.3% vs. 36.9% ± 13.8% with
a p-value of 0.123). Rituximab arm had a trend toward better survival compared with no
rituximab arm, but it was not statistically significant (66.7 vs. 55.8%) with a p-value of 0.584. Univariate Cox regression analysis showed significance only for age
with a p-value of 0. 001 with a hazard ratio of 10.3. ([Supplementary Material Table C], available in the online version) Twenty-four patients out of 55 did not receive
rituximab.
Fig. 1 Survival Chart Showing Progression Free Survival (PFS) and Overall Survival (OS)
at 8 years.
Discussion
BL is one of the most aggressive types of cancer and yet one of the most curable.
Mostly, BL originates from B cells in the follicular germinal center. BL is the first
lymphoma reported to be associated with HIV infection.[6] sBL in North America and Europe makes up 30 to 40% of childhood NHL and 1 to 5%
of adult NHL.[7] sBL attacks regions outside Africa, with an incidence of about 4 per million in
the United States.[8] In India, BL constitutes 1.8 to 3.4% of all adult NHL.[9] BL is a classic example of a human malignancy whose pathogenesis involves a specific
cellular genetic change characterized by a chromosomal translocation deregulating
expression of the c-myc oncogene, complemented in many cases by the action the EBV.[10]
This is the largest case series from India describing the treatment outcome of adolescent
and adult patients with BL. In our series, the median age at presentation was 42 years
(range: 14–81 years) with 11% patients above 65 years of age. These findings are consistent
with other case series.[3]
[11]
[12] In an Indian study including pediatric patients, BL constituted 3.5% of all NHL
and the median age at presentation was 22 years.[13] In this study, patients have a higher frequency of B-symptoms (65%) and advanced
disease compared to European studies.[14] In our study, bone marrow involvement was the most common extranodal site (30%)
followed by gastrointestinal tract (22%). CNS involvement in our study was 3% compared
to 15% in a study by Saleh et al.[14] Eleven patients (18%) had BL at presentation that was comparable with another study
by Mbulaiteye et al.[7]
Treatment success of BL resulted from the introduction of dose-intense multidrug chemotherapy,
prophylaxis of CNS disease, and improvements in supportive care. The management of
this aggressive lymphoma is a challenge in our resource-limited setting and the published
data from the Indian population is scarce. The frequently used regimens in adolescent
and adults based on pediatric protocols are CODOX-M/IVAC, the German BFM, and the
French Lymphome Malins B regimens. Other regimens used for adult BL include the HCVAD,
the Cancer and Leukemia Group B regimen, and the dose-adjusted EPOCH.[15]
[16]
[17] In our study, most of the young and fit patients received HCVAD ± R (63.5%) and
elderly patients with poor performance status received CHOP/CVP ± R (22%). The complete
response rate in the entire population was 62%, 2-year PFS of 58.3%, and a 2-year
OS of 61.6%. A study from India showed similar results with a complete response rate
of 63% with 2-year PFS of 56%.[18] Patients above 65 years of age and frail patient who received less intense therapy,
the 8-year OS was 46% that makes it an alternate option in elderly and frail patients.
Similar studies in BL in adult patients receiving CHOP-based chemotherapy had a 2-year
OS of 33 to 39%.[19]
[20] Patients who received HCVAD ± R chemotherapy had a complete response rate of 63%,
8-year PFS and OS of 60%. Review on treatment outcome in BL is given in [Table 4].[12]
[17]
[21]
[22]
[23]
[24]
[25] In this study, only a few patients received other chemotherapy regimens like CODOX-M/IVAC,
BFM, and Murphy's protocol to compare on survival. [Table 4] shows review of treatment outcome in BL.
Table 4
Review of treatment outcome in Burkitt's lymphoma
|
Reference
|
Regimen
|
n
|
Median age (years)
|
EFS/PFS
|
OS
|
|
Thomas et al[21]
|
Hyper-CVAD
|
26
|
58
|
3-year CCR 61%
|
3-year OS 49%
|
|
Thomas et al[17]
|
R-hyper CVAD
|
31
|
46
|
3-year EFS 80%
|
3-year OS 89%
|
|
Mead et al[12]
|
CODOX-M/IVAC
|
53
|
37
|
2-year EFS 64%
|
2-year OS 67%
|
|
Magrath et al[22]
|
CODOX-M/IVAC
|
41
|
25
|
2-year PFS 92%
|
NA
|
|
Lacasce et al[23]
|
CODOX-M/IVAC
|
14
|
47
|
2-year PFS 92%
|
NA
|
|
Patekar et al[24]
|
CODOX-M/IVAC
|
18
|
38
|
1-year PFS 76%
|
1-year OS 81.1%
|
|
Rizzieri et al[25]
|
CALGB regimen
|
105
|
44
|
2-year EFS 74%
|
3-year OS 58%
|
|
Present study
|
Hyper-CVAD ± R
|
35
|
43
|
8-year PFS 56.3%
|
8-year OS 60%
|
Abbreviations: CALGB, Cancer and Leukemia Group B; CCR, continuous complete remission;
CODOX-M/IVAC, cyclophosphamide, doxorubicin, vincristine, methotrexate, etoposide,
ifosfamide and cytarabine; EFS, event-free survival; hyper-CVAD, hyper-cyclophosphamide,
adriamycin, vincristine, prednisolone/ cyclophosphamide; OS, overall survival; PFS,
progression-free survival.
Limitations
The study period was January 2005 to December 2014, during this period molecular study
for BL was not available at our center. This is a retrospective data analysis. Patients
were not treated with uniform protocol as they received different protocols depending
on the time period the diagnosis was made.
Future Research Directions
Risk-adapted treatment that limits early and late drug toxicity ensuring optimal outcome
should be adopted for BL. In elderly patients, R DA EPOCH is the preferred choice
with better outcomes. Randomized control trials in Indian population are required
to better define the treatment for elderly patients with BL.
Conclusions
BL accounts for 1.57% of NHL above the age of 14 years with male preponderance. Intensive,
short-duration chemotherapy is the standard treatment. Treatment with HCVAD ± R gives
8-year PFS of 60% at 8 years. Treatment with R CHOP is an alternative option in elderly
frail populations with an 8-year OS of 46%.
