Beruflich Dokumente
Kultur Dokumente
Haematology
Manuscript Draft
Manuscript Number:
Title: Plasma phospholipid changes as a potential prognostic tool for
response to chemotherapy in non-Hodgkin lymphoma patients
Article Type: Articles (Original Research)
Keywords: plasma phospolipid profile, chemotherapy, Non-Hodgkin lymphoma
Manuscript
Zorica Cvetkovic1, #, MD, PhD, Maja Milosevic2,#,*, PhD, Bora Cvetkovic3, MD,
PhD, Romana Masnikosa4, PhD, Aleksandra Arsic5, PhD, Vesna Vucic5, PhD
Department of Urology, Clinical Hospital Center Zemun, Vukova 9, 11080 Belgrade, Serbia
Institute for Medical Research, Department for Nutrition and Metabolism, University of
Belgrade, Tadeusa Koscuska 1, 11129 Belgrade, Serbia
ABSTRACT
Background Limited studies have been performed to associate abnormal phospholipid (PL)
profile and disease activity in hematological malignancies. Altered PL metabolism has been
previously observed in non-Hodgkin lymphoma (NHL). The aim of his study was to evaluate
the levels of plasma PL fractions in NHL patients, in response to chemotherapy.
Methods Forty non-treated patients with NHL and 25 healthy individuals were recruited from
the Department of Hematology Clinical Hospital Center Zemun, Belgrade. Blood samples
from patients were taken before chemotherapy, after 3 cycles and after the end of the
treatment, and PL fractions were resolved by one-dimensional thin-layer chromatography. To
assess potential prognostic value of plasma PL profile, patients were divided according to
clinical outcome in 3 groups: complete remission (CR), stable disease (SD) and progression
(PG).
Findings In spite of significant differences between NHL patients and healthy controls, no
differences were found at baseline among patients dividing according to clinical outcome.
During and after chemotherapy important alterations in PL profile were observed: level of
total PLs and all PL fractions further decreased in patients with PG while in patients who
responded to therapy (CR, SD) PLs significantly increased.
Interpretation Results of our study suggest that monitoring changes of total PLs and PL
fractions during the therapy could be useful prognostic tool on the effects of therapy and
clinical outcome in patients with NHL.
Funding Ministry of Education, Science and Technological Development, Republic of Serbia
RESEARCH IN CONTEXT
to the clinical stage, and aggressiveness of NHL. In particular, marked changes were found in
baseline values of phospholipid fractions when the patients were divided based on the clinical
outcome after completed chemotherapy.
Implications of all the available evidence
The present study showed that PL profile in NHL patients is associated with the response to
chemotherapy, depending on aggressiveness of disease, clinical stage and clinical outcome.
These findings reveal phospolipids as potential biomarker in prognosis and treatment of NHL.
INTRODUCTION
Lipids are a diverse class of essential molecules with the role in cellular structure,
energy storage and signaling. De novo biosynthesis of fatty acids (FA) and cholesterol is
restricted to liver, adipose and lactating breast tissue, while the rest of mammalian tissues
uptake lipids from the bloodstream. Development and progression of most tumors is followed
by de novo FA and cholesterol synthesis, but also by the elevated uptake of dietary lipids
from the circulation. As a consequence, tumor influences systemic lipid homeostasis that is
reflected in body fluids.1 Phospholipids (PLs) are key components of cellular membranes and
important bioactive molecules. Characteristics of tumor tissue compared with normal tissue
are elevated levels of cell membrane phosphatidylcholine (PC) and phosphatidyletanolamine
(PE). Increased FA saturation of PL in cancer cell membranes alter fluidity and signal
transduction, affecting cells chemotherapy resistance and protection from oxidative damage.1
It is known that metabolism of lipids is frequently altered in various diseases, but recent
evidence indicate that lipid-related genes may link cancer with inflammatory and metabolic
diseases.2 Lipidomic studies have demonstrated that the risk,3 existence4 and progression5 of
the same kind of cancer may be correlated with altered level of particular tissue and plasma
PL and fatty acids. The most recent data from a large cohort study found an association
between the risk of tree frequent type of cancer and plasma levels of LPC (C18:0) and PC
(C30:0).3 Differences in expression pattern of tissue/plasma PL level showed good prognostic
performance in distinguishing breast cancer from benign tumors as well as uterine fibroids
from cervical cancer.4
Non-Hodgkin lymphomas (NHL), a heterogeneous group of hematologic malignancies, show
rapid increase in the incidence over the past few decades. Etiology is still unclear although
some autoimmune disorders, infectious agents and lifestyle factors including diet, appear to
play a role in pathogenesis of NHL.6 Identification of altered PL in human lymphoma and
animal models revealed the link between lipid signature and specific oncogens, such as v-mic
avian myelocytomatosis viral oncogen (MYC).7 Survival rate of the most common subtypes
of NHL: follicular lymphoma and diffuse large B-cell lymphoma have been improved in
European region as a result of early diagnosis and advances in treatment.8
We have previously reported significantly lower levels of total cholesterol, HDLcholesterol and total PL in serum of patients with NHL.9 Furthermore, altered plasma FA
profile in NHL patients was linked to clinical stage and aggressiveness of the disease,10 as
Analytical Methods
Venous blood samples were drawn after an overnight fast, prior to starting
chemotherapy (baseline), after third cycle of chemotherapy regimen (middle) and after the
completion of therapy (end). Serum triglyceride, total cholesterol and HDL cholesterol
concentrations were assayed by the automated enzymatic methods (Roche, Basel,
Switzerland). LDL cholesterol was estimated using the Friedewald formula.14 The total PL
concentration was determined by the Zilversmit method.15 Plasma lipids were extracted with
chloroformmethanol mixture (2:1 v/v) as we previously described.16 The PL fraction was
isolated from the extracted lipids by one-dimensional thin-layer chromatography in a neutral
solvent system (petrol ether -diethyl ether - acetic acid; 87:12:1 v/v) on Silica Gel GF plates
(Merck, Darmstadt, Germany). Four fractions of PL were detected in plasma:
lysophosphatidylcholine (LPC), sphingomyelin (SM), phosphatidylcholine (PC) and
phosphatidyletanolamin (PE).
Statistical analysis
All the results are expressed as the mean SD. Normality was tested using the
Kolmogorov-Smirnov test. Since all variables showed normal distribution, one-way ANOVA,
followed by the Tukey post hoc test, and the Student t-test for the comparisons between two
groups were used. In addition, paired Student t-test was performed to follow the time course
changes of the parameters in the same patient at the beginning, in the middle and after the
chemotherapy. The level of significance was set at p0.05.
Role of the funding source
The funder had no involvement in the study design, data collection, analysis and
interpretation, writing of the manuscript, or decision to submit for publication. ZC, MM and
VV had full access to all data in the study. MM and ZC had the final responsibility for the
decision to submit for publication.
RESULTS
Blood lipid and PL profile of patients with NHL, before, in the middle and after
chemotherapy are presented in Table 1. Untreated NHL patients have significantly lower
concentrations of plasma lipid and PL fractions, when compared with healthy individuals,
with an exception of triglycerides which was similar in both groups. During and after
chemotherapy regimen, all measured lipids gradually increased in NHL patients, but remained
significantly lower when compared with the control group, except for HDL-C level which
almost achieved value of the controls.
When the patients were divided according to aggressiveness of NHL into indolent (I),
aggressive (A), and very aggressive (VA) NHL, we found a significantly lower concentration
of total PL, PC and LPC in the A (p<0.05) and VA group (p<0.01) when compared with the
indolent group at baseline (Table 2). In all 3 groups, level of total PLs and most of PL
fractions showed an increasing trend during and after the end of chemotherapy, but these
changes were nonsignificant. Thus the significant differences between the groups A and I
were lost after the 3 cycles of chemotherapy, and after completion of the therapy between the
VA and I group. Furthermore, after receiving the complete therapy, only the patients with
indolent NHL reached the level of all PL comparable to the healthy subjects.
Patients were also grouped according to clinical stage of NHL (CS I-IV). There were no
significant differences between the groups at baseline, in spite of a decreasing trend of
concentrations of all PLs from CS I to CS IV (Table 3). Interestingly, the treatment led to
significant differences of all PLs between CS II and CS IV, as well as of total PL, LPC and
SM between CS III and CS IV, while no differences were found between CS I and any other
group. However, the most notably differences were found when the patients were divided
according to the response to therapy in 3 groups: complete remission (CR), stable disease
(SD) and progression (PG).
As it can be seen in Table 4, the patients in CR and SD showed increasing trend in PL
profiles in response to chemotherapy, while the PG group showed the opposite trend. Patients
in the CR group showed a marked elevation of all parameters except PE, attaining the PL
profiles statistically equivalent with those from healthy subjects. On the contrary, PG group
displayed a gradual decrease of all PLs, except PE, at the end of chemotherapy. Calculated
PC/LPC ratio, as a parameter of inflammation, inversely followed observed trend in PL
profile and declines in CR and SD group, while it increases in PG.
When we compared differences among the 3 groups at the same time points by one way
ANOVA, we found no baseline differences, but significant differences between CR versus SD
and especially PG groups during and after the therapy (Table 4, points 2 and 3). In addition,
the levels of all PLs, except PE, were significantly lower at the end of therapy in patients with
PG than in SD.
DISCUSSION
In the present study PL profile of NHL patients was for the first time associated with the
response to chemotherapy, depending on aggressiveness of disease, clinical stage and clinical
outcome.
Observed lipid profiles in NHL patients before during and after chemotherapy were
consistent with previous studies.9,17 Our results also revealed lower level of all PL fractions in
NHL patients than in healthy subjects, but even lower baseline level of total PL, PC and LPC
in more aggressive types of disease than in indolent NHL. On the other hand, no differences
were found at baseline according to clinical stadium, but an increase of all PL fractions was
detected during the therapy in CS II and III, unlike CS IV. Low plasma level of PL had been
previously correlated with two-fold shorter survival in cancer patients.18
Although a recent lipidomic research showed altered level of different PL species in
various cancers, level of LPC in biofluids appears to be more significant prognostic biomarker
than total PL.19 LPC is a common plasma constituent and its decrease have been reported in
many cancer types, including lymphoma.20 Reduced LPC level in blood has also been linked
to activated inflammatory status in cancer, and may be result of a higher consumption of LPC
by cancer cells, as a source of FA or signaling molecules which induce tumor growth, such as
lysophosphatidic acid (LPA). Elevated expression of autotaxin (ATX), a secreted
lysophospholipase D, found in several cancer types, may lower LPC by its conversion into
LPA.21 In addition, LPC-acyltransferase 1, enzyme which converts LPC into PC, is also
overexpressed in cancers.22
In our study LPC was significantly higher in indolent NHL than in the aggressive forms,
and its concentration varied depending on the clinical outcome. During the therapy LPC level
markedly increased in patients with complete remission, slightly increased in stable disease,
but further decreased in patients with progression of NHL. The same trend has been observed
for the most abundant PL in plasma PC, as well as for the other PL fractions SM and PE. On
the other hand, the PC/LPC ratio has shown the opposite trend. We have also found similar
baseline levels of PC and LPC in all patients, suggesting that these parameters are not useful
as prognostic factors in NHL. However, monitoring the changes of PC, LPC and their ratio
during the therapy can predict the response to therapy and clinical outcome in patients with
NHL.
The other PL classes were also changed. Sphingolipids (SL) and
phosphatidyletanolamine (PE) are constituents of lipoproteins but their role in circulation is
not clear.23 Some studies showed decreased SM levels in glioma cells24 and lower serum SM
concentrations were also found in hepatitis, cirrhosis and hepatocellular carcinoma.25 Eleveted
concentration of plasma SL are associated with the development of metabolic syndrome26 and
some types of cancers.27 Serum PE level was found to be potential diagnostic biomarker to
differentiate benign from malignant lung28 and prostate tumors.29
Some of chemotherapeutics (doxorubicin, vincristine, fludarabine and rituximab) used
in our study can affect lipid metabolism via ceramide accumulation.30 The successful
treatment of B-cell-derived lymphoma by rituximab is showed to be the result of increase in
acid-sphingomyelinase activity and ceramide generation.31 Ceramides initiate apoptosis via
activation of caspases, block activation of phopholipase D and survival signals initiated by
phosphatide. Nevertheless, almost 1/3 of analyzed NHL patients showed disease progression.
Resistance to chemotherapy depends largely on the effects of various growth factors,
including LPA. Overexpression of ATX and LPA receptors has been proposed as possible
mechanism of resistance to chemo- and radiotherapy. LPA activates phopholipase D and
phosphatidylinositol 3-kinase, which increase production of sphingosine-1-phosphate and
decrease production of ceramide, shifting sphingolipid rheostat toward cells survival.32
An important limitation of this study is the lack of fatty acid composition of PL
fractions. A recent prospective study has shown that higher plasma LPC C18:0 was constantly
related to lower risk of common cancers, while higher PC C30:0 was associated with
increased cancer risk.3 Many other chronic diseases are also connected with PLs and their FA
profiles. For instance, phosphatidylinositol (PI), phosphatidylglycerol (PG), and PE were
found to be positively associated with prediabetes and type-2 diabetes33 while PE
glycerophospholipid (C36:2), and the diacylglycerol species (C36:2, C34:0) were linked with
metabolic syndrome risk.34 Based on these data, it could be expected that PLs with certain FA
composition have even stronger prognostic potential and future research should go in this
direction.
In conclusion, PLs and their fractions may be promising in monitoring the efficacy of
the treatment and predicting clinical outcome in NHL patients. Further studies on a larger
number of patients, which would also investigate possible influence of cofounding factors
(gender, age and body mass index), are needed to establish an useful prognostic value of
plasma PLs in patients with NHL.
Contributors
VV and ZC designed the study. ZC and BC recruited patients and collected blood samples.
AA, MM and RM performed experiments and analyzed data. VV did the statistical analysis.
MM searched literature and drafted the manuscript. VV and RM did substantial
reorganization and editing of the manuscript. All authors have approved final version of the
manuscript.
Declaration of interests
The authors declare that they have no conflict of interest.
Acknowledgments
This work was supported by the Projects III 41030 and III 41014 financed by the
Ministry of Education, Science and Technological Development of the Republic of Serbia.
REFERENCES
1
2
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
NHL (n=40)
Time point
Control (n=25)
3
TC
3.650.72***
3.880.93***
4.091.14***
5.290.63
HDL-C
1.050.16***
1.070.26*
1.110.32
1.240.27
LDL-C
2.290.62***
2.450.70***
2.640.85***
3.7740.6
TG
1.530.37
1.650.34***
1.690.36***
1.380.26
PL
2.230.33***
2.330.40***
2.430.52***
2.900.39
PC
1.050.14***
1.100.18***
1.110.29***
1.360.18
LPC
0.340.06***
0.360.08***
0.390.12**
0.490.10
SM
0.570.09***
0.590.11***
0.620.13**
0.730.12
PE
0.270.05**
0.280.09
0.280.06*
0.320.07
PC/LPC
3.140.22**
3.040.24
3.050.30
2.860.56
Table 2 Concentrations (mmol/l) of total plasma PL and PL fraction in patients divided according to
aggressiveness of NHL
PL
PC
LPC
SM
PE
PC/LPC
I (n=15)
A (n=17)
VA (n=8)
2.430.36
2.150.21*
2.030.30*
2.530.39
2.280.37
2.090.32*
2.610.54
2.340.49
2.230.46
1.130.16
1.02010*
0.940.11**
1.190.18
1.070.16
0.980.13*
1.230.25
1.110.22
1.000.12
0.380.06
0.330.05*
0.300.05**
0.400.08
0.350.08
0.310.07*
0.430.14
0.360.10
0.340.09
0.610.09
0.550.08
0.540.08
0.640.11
0.580.11
0.540.08
0.670.12
0.600.15
0.570.10
0.280.06
0.260.04
0.250.07
0.290.04
0.270.04
0.310.18
0.300.06
0.270.05
0.260.06
2.990.29
3.120.25
3.200.21
2.970.23
3.140.44
3.180.30
2.920.40
3.100.31
3.010.83
I indolent NHL, A aggressive, VA very aggressive, before (1), in the middle (2) at the end of
chemotherapy (3)
*p<0.05, **p<0.01, when compared with the group I.
Table 3 Concentrations (mmol/l) of total plasma PL and PL fraction in patients divided according to
clinical stage (CS I-IV) of NHL
CS I (n=4)
CS II (N=11)
CS III (n=11)
CS IV (n=13)
2.410.54
2.310.22
2.240.23
2.100.38
2.290.52
2.620.35**
2.410.34*
2.040.24
2.170.60
2.750.40*
2.540.62
2.100.17
1.040.15
1.090.09
1.070.13
0.990.18
1.070.22
1.230.17**
1.130.14
0.970.12
1.170.23
1.240.22*
1.180.29
0.990.09
0.370.10
0.370.05
0.330.04
0.320.07
0.360.11
0.430.07***
0.370.07*
0.300.05
0.390.25
0.460.09*
0.390.12
0.310.03
0.630.14
0.590.07
0.570.05
0.530.10
0.590.14
0.670.11**
0.620.09**
0.510.05
0.550.15
0.690.10*
0.660.17
0.550.05
0.310.09
0.280.05
0.260.03
0.250.06
0.270.06
0.340.14
0.290.05
0.250.03
0.250.08
0.300.05*
0.290.07
0.250.03
PC/LPC 1
2.980.31
3.200.14
3.150.23
2.910.43
3.040.29
2.890.22*
3.050.23
3.290.44
3.091.16
2.770.46*
3.070.33
3.170.19
PL
PC
LPC
SM
PE
Table 4 Plasma phospholipid profile (mmol/l) after end of chemotherapy in NHL patients according to
their responce to therapy: complete remission (CR), stable disease (SD) progression (PG)
PL
CR (n=11)
SD (n=13)
PG (n=13)
2.370.30
2.180.29
2.240.37
2.750.26
2.300.34**
2.100.23***
a
1.940.16***### a
3.000.20
2.460.46**
1.080.10
1.040.15
1.050.17
1.290.13
1.080.14**
0.990.10***
1.330.21
1.170.22
0.950.08***### aa
0.370.06
0.340.06
0.340.06
0.460.05
0.350.06***
0.310.04***
0.520.09a
0.380.08***
0.280.04***## aa
0.620.08
0.550.08
0.570.10
0.700.07
0.580.10**
0.530.06***
0.750.06
0.640.12* a
0.500.05***### a
0.290.05
0.240.04
0.280.06
0.350.14
0.270.05
0.260.04*
0.340.04
0.270.05***
0.230.03***
PC/LPC 1
3.010.21
3.220.19*
3.130.19
3.070.23**
3.130.20***
3.040.17**
3.310.26***## aa
PC
LPC
SM
PE
2
3
2.820.98
2.750.10
aa
*p<0.05, **p<0.01, ***p<0.001 when compared with CR. ##p<0.01, ###p<0.001 when compared with
SD (ANOVA followed by the Tukey post hoc test).
a
aa
p<0.05, p<0.01, p<0.001 when compared with baseline values (time point 1), p<0.05,
p<0.01 when compared with the middle of chemotherapy (time point 2), evaluated by paired t-test.