Key words
vitamin C - ascorbic acid - cancer - chemotherapy - ipilimumab - scurvy
Introduction
An estimated 3.2 million new cases of cancer and 1.7 million deaths from cancer occurred
in Europe in 2008 [1]. The fact that only 5–10% of all cancer cases are due to genetic defects and that
the remaining 90–95% are due to lifestyle factors (including smoking, alcohol, physical
inactivity, obesity and sun exposure), nutrition, infections and environmental pollutants
provide major opportunities for preventing cancer [1]. Malignant melanoma is a tumor derived from melanocytes, manifesting mainly on the
skin, and in rare cases, from mucous membranes. The tumor-specific 10-year survival
for patients with primary melanomas without any evidence of metastasis is 75–85%,
decreasing to 20–40% for those patients with clinically apparent regional lymph node
metastases. Distant organ metastases (stage IV disease) have a poor prognosis with
a median survival in untreated patients of 6–9 months [2].
Several studies have shown that plasma ascorbate levels are decreased in patients
with malignancies, such as hepatocellular carcinoma [3], lung cancer [4], colorectal cancer [5] and prostate cancer [6].
A basic biochemical role for ascorbate is to accelerate hydroxylation reactions in
a number of biosynthetic pathways. In many of these reactions, ascorbate directly
or indirectly provides electrons to enzymes that require prosthetic metal ions in
a reduced form to achieve full enzymatic activity. Ascorbate is required in the synthesis
of carnitine from lysine, collagen and neurotransmitter synthesis, cytochrome P-450
activity, cholesterol metabolism, detoxification of exogenous compounds, and as an
antioxidant [7]. Moreover, ascorbate is essential for immunoglobulin synthesis. In cell-mediated
immunity, immunocompetence is exercised overwhelmingly by lymphocytes, which contain
high concentrations of ascorbate relative to other cells. In addition, ascorbate is
required for active phagocytosis and enhances interferon production [7]. In line, impairment of lymphocyte activation, and macrophage and neutrophil mobilization
were reported in scorbutic animals [9].
Aim of the study: Considering the importance of adequate ascorbate supply for hundreds of biochemical
processes and a functional immune system, we asked whether plasma ascorbate levels
correlated with disease stage in patients afflicted with malignant melanoma and whether
intravenous systemic therapy (polychemotherapy or immunotherapy) in stage IV melanoma
patients influenced the plasma ascorbate level.
Materials and Methods
Ethics statement
The use of human plasma samples was approved by the medical ethical committee of the
University of Tuebingen, Germany, and was performed in accordance with the Declaration
of Helsinki Principles. The plasma samples were gathered during routine blood withdrawals.
Every patient signed an informed consent form prior to blood withdrawal allowing the
determination of plasma ascorbate level for research purposes.
Recruitment of patients and controls
Patients aged 18 years and older included in this study. Patients with artificial
ascorbate supplementation (e. g. taking ascorbate- or multi-vitamin pills) were excluded
from this study. The control cohort consisted of healthy volunteering employees (nurses,
dermatologists and laboratory staff) of the Department of Dermatology, University
of Tuebingen, Germany, aged 18 years and older and was stratified by the exclusion
criteria stated above. The control cohort and the stage I and stage II melanoma patients
had an ECOG Performance Status grade 0. The stage III melanoma patients had an ECOG
Performance Status grade 0 or 1, and the stage IV melanoma patients grade 3 or 4.
HPLC ascorbate measurements
For the quantification of ascorbate in EDTA-plasma, samples were immediately mixed
with an equal amount of 5% perchloric acid (Sigma, Munich, Germany) and stored at
−80°C. The time between blood withdrawal and snap-freezing was less than 10 min. For
analysis of plasma ascorbate content, samples were thawed and centrifuged at 13 000
× g for 5 min at 4°C. For reduction of DHA and determination of TAA, the supernatant
was mixed with 0.15 M aqueous Tris(2-carboxyethyl)phosphine (TCEP, ABCR, Karlsruhe,
Germany) solution (v/v: 2/1) and incubated for 5 min. Afterwards, the mixture was
centrifuged at 13 000 × g for 5 min at 4°C and transferred into light-protected micro-vials
for chromatographic analysis. As external standards, 2 aqueous ascorbic acid solutions
(50 µM and 100 µM) were prepared, diluted with an equal amount of perchloric acid
(5%), mixed with 0.15 M aqueous TCEP solution (v/v: 2/1) incubated for 5 min. The
mixture was centrifuged at 13 000 × g for 5 min at 4°C and transferred into light-protected
micro-vials for chromatographic analysis. All analytical procedures were performed
on ice and under dimmed light to avoid ascorbic acid degradation. For HPLC analysis,
samples were placed in a cooled autosampler (Varian, Darmstadt, Germany). Chromatography
was carried out on a reversed-phase column (Reprosil-Pur 120 C18 AQ 5 µM; Trentec,
Gerlingen, Germany) with a mobile phase consisting of 5 mM aqueous sodium phosphate
buffer, pH 2.5 at a flow rate of 1 ml/min. Detection was performed by using an electrochemical
detector (ESA Detektor Coulochem II, Chelmsford, United Kingdom) and a Model 5011
high-sensitivity analytical cell (ESA) at −300 mV (E1, upstream) and +300 mV (E2,
downstream). Chromatograms were recorded and analyzed using Star chromatography workstation
software version 5.31 (Varian, Darmstadt, Germany).
Statistical analysis
Statistical analysis of plasma ascorbate level was performed by using Student’s t-test.
P-values<0.05 were considered as statistically significant.
Results
Plasma ascorbate is decreased in more advanced stages of melanoma and negatively correlates
with plasma S100 and LDH levels
We determined total ascorbate (TAA) levels in plasma of healthy control individuals
(n=31, mean age: 47.3 years, TAA: 64.86 µM [26.68 µM-113.64 µM; SD: 1.11 µM]) and
in 126 melanoma patients (stage I: n=30, mean age: 51 years, TAA: 59.95 µM [33.80 µM–106.25 µM;
SD: 1.0 µM]; stage II: n=30, mean age: 46.8 years, TAA: 58.85 µM [28.83 µM–88.06 µM;
SD: 1.20 µM]; stage III: n=32, mean age: 48.6 years, TAA: 57.27 µM [27.63 µM–80.11 µM;
SD: 1.17 µM]; stage IV: n=34, mean age: 51.1 years, TAA: 47.16 µM [4.82 µM–73.99 µM;
SD: 1.0 µM]) by HPLC ([Table 1]). Plasma TAA levels constantly decreased from stage I to stage III, but the difference
was not significant when compared to the healthy controls. However, in stage IV patients
the ascorbate plasma levels were significantly reduced by 27.3% when compared to healthy
individuals (p=0.0001, t-test). The ascorbate plasma levels of stage IV patients were
also significantly lower than those of stages I–III patients (p=0.0006 for stage I
vs. stage IV; p=0.0041 for stage II vs. stage IV; p=0.0044 for stage III vs. stage
IV; t-test) ([Table 1]). Plasma ascorbate levels negatively correlated with S100 and lactate dehydrogenase
(LDH)-values (both prognostic markers for malignant melanoma) in stage IV patients:
The lowest plasma ascorbate concentrations were measured in melanoma patients with
pathologically elevated (>0.1 µg/l) S100-values (mean TAA: 44.04 µM [4.82 µM–73.99 µM;
SD: 0.94 µM]; 32.1% reduction compared to controls; p=0.0002, t-test) and elevated
(>250 U/l) LDH-values (mean TAA: 37.76 µM [4.82 µM–64.28 µM; SD: 0.92 µM]; 41.8% reduction
compared to controls; p=0.0001, t-test) ([Table 1]).
Table 1 Plasma total ascorbate levels of 126 stage I–IV melanoma patients and 31 controls.
|
cohorts
|
n
|
mean age
|
TAA [µM]
|
% reduction in TAA
|
p-value, t-test
|
|
control
|
31
|
47.3
|
64.86
|
|
|
|
stage I
|
30
|
51
|
59.95
|
7.6
|
0.253
|
|
stage II
|
30
|
46.8
|
58.85
|
9.3
|
0.196
|
|
stage III
|
32
|
48.6
|
57.27
|
11.7
|
0.069
|
|
stage IV
|
34
|
51.1
|
47.16
|
27.3
|
0.0001
|
|
stage IV S100<0.1 µg/l
|
12
|
52.8
|
51.78
|
20.2
|
0.028
|
|
stage IV S100>0.1 µg/l
|
21
|
50.9
|
44.04
|
32.1
|
0.0002
|
|
stage IV LDH<250 U/l
|
21
|
51.8
|
52.05
|
19.8
|
0.006
|
|
stage IV LDH>250 U/l
|
12
|
51.1
|
37.76
|
41.8
|
0.0001
|
TAA: total ascorbic acid; LDH: lactate dehydrogenase
Plasma ascorbate decreases after poly-chemotherapy and immunotherapy
Second, we asked whether plasma ascorbate levels were influenced by systemic intravenous
therapies performed on stage IV melanoma patients. To this end, plasma ascorbate levels
were determined in 9 patients directly before and 24 h after either polychemotherapy
(carboplatin+paclitaxel) or immunotherapy targeting CTLA4 (ipilimumab) [9]
[10] ([Table 2]). We detected significantly reduced plasma ascorbate levels 24h after therapy (48.7±17.4 µM
vs. 43.0±14.7 µM; p<0.05, t-test) ([Table 2]). We could not find a significant difference of the reduction of plasma ascorbate
levels caused by poly-chemotherapy when compared to immunotherapy.
Table 2 Plasma total ascorbate levels before and after polychemo- and immunotherapy in stage
IV melanoma patients.
|
Patient
|
Treatment
|
TAA [µM] pre
|
TAA [µM] post
|
% reduction in TAA
|
p-value, t-test
|
|
1
|
Carbotax
|
41.1
|
36.2
|
11.9
|
|
|
2
|
Carbotax
|
8.5
|
6.9
|
18.1
|
|
|
3
|
Carbotax
|
53.9
|
46.0
|
14.7
|
|
|
4
|
Carbotax
|
64.3
|
45.4
|
29.4
|
|
|
5
|
Carbotax
|
46.9
|
42.2
|
10.1
|
|
|
1
|
Ipilimumab
|
74.0
|
64.4
|
13.0
|
|
|
2
|
Ipilimumab
|
56.4
|
51.6
|
8.4
|
|
|
3
|
Ipilimumab
|
52.3
|
48.8
|
6.6
|
|
|
4
|
Ipilimumab
|
41.1
|
45.2
|
−9.9
|
|
|
mean
|
48.7
|
43.0
|
11.4
|
<0.05
|
TAA: total ascorbic acid; pre: before therapy; post: 24 h after therapy; Carbotax:
carboplatin+paclitaxel
Discussion
Critical steps in the determination of plasma ascorbate levels
In this study we show for the first time (to the best of our knowledge) that plasma
ascorbate levels are reduced in stage IV melanoma patients when compared to healthy
age-matched controls. Comparing the mean plasma ascorbate level of our healthy control
cohort (64.86 µM), we find a discrepancy in comparison to plasma ascorbate concentrations
in the literature. In the European Prospective Investigation into Cancer and Nutrition
(EPIC-Norfolk prospective study), the mean ascorbate serum concentrations in the different
cohorts were 20.8–72.6 µM with a calculated mean of 47.3 µM (the exact mean was not
communicated in the study) [11]. Secondly, in the more recent European Prospective Investigation into Cancer and
Nutrition (EPIC-EURGAST), the mean ascorbate serum concentration in controls was 41.5 µM
[12]. However, in the latter 2 studies the processing of the blood samples for the determination
of ascorbate levels was not comparable to the conditions used in the present study.
According to the methods section of the 2 studies, the authors did not enforce a quick
handling and freezing of the blood samples. Considering the importance of a quick
processing (stabilization and subsequent freezing in liquid nitrogen), which is indispensable
due to the instability of ascorbic acid, and which we performed within the short time
of less than 10 min after blood withdrawal for all samples, this obvious difference
in sample handling most likely accounts for the difference of ascorbate concentrations
on our control group when compared to the mean values of the latter 2 studies. In
line with this assumption, in the second National Health and Nutrition Examination
Survey, the median serum ascorbate concentrations were 49.4 µM for men and 64.2 µM
for women [13]. In the latter study, the mean ascorbate concentration was comparable to our control
group, and the blood samples were put on dry ice after stabilization in the mobile
examination centers without further storage at 7°C over night. Thus, the quick stabilization
and constant freezing (on dry ice) is comparable to the handling of the samples in
our study and most likely accounts for the similar results obtained in the control
samples. The lack of significant reduction of ascorbate plasma levels in stage I–III
patients was obviously related to tumor burden. Stage I and II melanoma patients are
completely tumor-free after resection of the primary melanoma, and stage III patients
after adequate surgical treatment most of the times as well. Only in stage IV there
is a metastatic tumor burden with all its consequences.
Impact of high plasma ascorbate levels for cancer development
Ascorbate-rich diets have protective effects on cancer development and other age-related
diseases [14]. A recent meta-analysis of 8 case-control studies found a significant preventive
effect on cervical neoplasms for increased intake of ascorbate [15]. Further, large epidemiological studies demonstrated a significantly increased risk
of dying from cancer for adults [13] and for men [11] with low serum ascorbate concentrations. The prospective “European Prospective Investigation
into Cancer and Nutrition” study revealed that higher plasma vitamin C levels were
associated with a decreased risk of gastric cancer, independent of the particular
gastric cancer anatomical subsite or histological subtype [12].
Impact of low plasma ascorbate levels for cancer patients
Previous clinical studies demonstrated reduced ascorbate levels in non-melanoma cancer
patients [4]
[16] sometimes even leading to scurvy [17]. In the latter case series the authors stressed the fact that scurvy should be considered
in patients with cancer because of the high incidence of malnutrition caused by chemotherapy,
cachexia caused by the disease, and other factors that might lead to an unbalanced
dietary intake such as exclusive parenteral nutrition, depression, impaired taste,
dysphagia, and abdominal pain. Further, weakness, anorexia, and depression are common
in scurvy but also in patients with cancer. The authors concluded that clinicians
should suspect ascorbate deficiency when a patient has haemorrhagic features without
a clear explanation [17].
The importance of ascorbate for innumerable biochemical processes in the human organism
is beyond dispute. Since humans and other primates lack the enzyme L-gulonolactone
oxidase required for ascorbate synthesis, an adequate oral uptake is vital [7]. Considering the dependency of the immune system on adequate ascorbate supply [8] and the importance of the immune system for endogenous tumor control, and that,
in addition, stage IV cancer patients frequently suffer from infections caused by
therapy- or disease-induced neutrophilia, the additional ascorbate supplementation
in physiological doses could be taken into consideration.
Impact of anticancer therapies on plasma ascorbate levels
Our data showed that stage IV melanoma patients are ascorbate-deficient, and that
systemic therapies commonly used on such patients (chemotherapy, immunotherapy) further
reduced the plasma ascorbate levels. However, the low number of patients in this cohort
(n=9) limits the generalizability of this finding to all stage IV melanoma patients
undergoing systemic therapies in spite of its statistical significance. A similar
therapeutic phenomenon was described by Marcus and colleagues, who demonstrated a
severe ascorbate deficiency in 11 cancer patients (with metastatic melanoma, hypernephroma,
and colon carcinoma), occurring as the result of adoptive immunotherapy with high-dose
interleukin 2 and lymphokine-activated killer cells [8]. Considering that so far no study has shown negative effects of ascorbate supplementation
in cancer patients, our results suggest that especially late-stage melanoma patients
(stage IV) might benefit from ascorbate supplementation.
Potential anticancer mechanisms of ascorbate on melanoma cells
A number of recent pre-clinical reports have demonstrated a decreased melanoma growth
(both after s. c. and intraperitoneal injection) in mice unable to synthesize ascorbic
acid (gulonolactone oxidase knockout mice) upon ascorbate supplementation [18]
[19]. In the latter 2 studies, ascorbate supplementation was accompanied by decreased
serum pro-inflammatory cytokine (IL-6, IL-1β) levels as well as enhanced encapsulation
of tumors. IL-6, secreted by melanoma cells, promotes melanoma growth by inhibition
of apoptosis and induces tumor angiogenesis. Further, an increased serum concentration
of IL-6 correlates with a worse prognosis in melanoma patients [20]. This obvious impact of ascorbate on the immune response is in line with its crucial
role in immunoglobulin synthesis. Moreover, lymphocytes contain high concentrations
of ascorbate, and ascorbate is required for phagocytosis and enhances interferon production
[7]. In addition, scorbutic animals show an impairment of lymphocyte activation and
macrophage and neutrophil mobilization [8]. Together, for the immune response, ascorbate seems to be a double-edged sword functioning
both as pro- and anti-inflammatory stimulus that inhibits tumor growth. On the cellular
level, growth inhibition of melanoma cells was driven by cell cycle arrest and caspase-3
activation with subsequent apoptosis induction [21]. Further, ascorbate reduced the expression of the transferrin receptor in melanoma
cells, resulting in a down-regulation of cellular iron uptake followed by apoptosis
induction [22]. Due to recent pharmacokinetic data showing that intravenous administration of ascorbate
results in high plasma levels not achievable by oral uptake and in a significant flux
of H2O2 to tumors, the use of i. v. ascorbate in pharmacological doses in the treatment of
cancer seems plausible and is currently being investigated in clinical trials and
in experimental studies [23].
Conclusions
We were able to demonstrate that plasma ascorbate levels were significantly lower
in stage IV melanoma patients. Systemic therapy on stage IV patients further reduced
plasma ascorbate levels. Considering the importance of adequate ascorbate supply for
biochemical processes and host defense, our results suggest that late-stage melanoma
patients might benefit from physiological ascorbate supplementation.
