Could a citrus keep the haematologist away?


Anaemia has long been recognized as one of the most important prognostic factors in chronic lymphocytic leukaemia (CLL) (Binet et al, 1981). Many mechanisms can cause anaemia in CLL including bone marrow infiltration, auto-immunity, cytotoxic therapy, hypersplenism, inflammation, iron deficiency and poor nutritional status (Mauro et al, 2002). Among nutritional causes of anaemia, vitamin B9 and vitamin B12 deficiencies are often sought, whereas vitamin C deficiency (hypovitaminosis C) is underestimated (Fain, 2004). Vitamin C, also known as ascorbic acid (AA), is one of the four main antioxidants (AA, vitamin E, selenium and β-carotene) available in human alimentation (Fain, 2004). The pathophysiology of CLL may involve oxidative stress (Sarmento-Ribeiro et al, 2012). Because of its role in preventing anaemia and its ability to degrade free radicals and oxidants, we hypothesized that hypovitaminosis C level could be associated with Binet stage C. Herein, we report a single-centre study comparing the blood level of vitamin C in patients with Binet stage A and Binet stage C CLL.

Between June 2012 and November 2012, we performed a prospective exploratory study of the vitamin C plasma level in patients followed for CLL at the Department of Hematology of University Hospital of Tours. In order to compare vitamin C plasma level between low and high burden of disease, we randomly selected 40 patients with Binet stage A CLL and 40 patients with Binet stage C CLL (Binet et al, 1981). The only exclusion criterion was ongoing vitamin C supplementation. The following baseline demographical and clinical data were recorded for all of the study patients: age, sex, weight, size, performance status and date of CLL diagnosis. The following blood tests were performed in patients who provided an informed consent: complete blood cell count, reticulocytes, aspartate aminotransferase, serum creatinine level, thyroid-stimulating hormone (TSH), C-reactive protein (CRP), serum iron level, transferrin saturation, ferritin, albumin, folic acid, vitamin B12 and vitamin C serum levels. To protect AA from light and air alteration, blood samples were collected in tubes encased in foil. Results are expressed as mean and confidence intervals. Wilcoxon's test was used to test the comparisons between the two groups. A P value of less than 0·05 was considered significant.

Baseline data and biological test results of the patients are presented in Table 1. Stage C patients had lower haemoglobin level than stage A patients (107 g/l, vs. 134 g/l, respectively, P < 0·00005). Anaemia was mostly normocytic, normochromic and hyporegenerative. Platelet count was also lower in stage C patients than in stage A patients (111 × 109/l, vs. 200 × 109/l, respectively, P < 0·00005). Thyroid, liver and renal functions, as well as vitamin B9 and serum iron levels, were normal in both groups. CRP and ferritin levels were significantly higher in patients with Stage C (10·0 mg/l, vs. 3·0 mg/l, and 431 μg/l, vs. 117 μg/l, respectively, P < 0·05 for both comparisons). There was no statistical difference in Vitamin B12 level between the two groups. Vitamin C level was significantly higher in stage A patients than in stage C patients (58·7 μmol/l, vs. 37·7 μmol/l, P < 0·00005). In total, 22 CLL patients (27·5%) had hypovitaminosis C including three stage A patients (7·5%) and 19 stage C patients (47·5%).

Table 1. Characteristics of Stage A and Stage C CLL patients included in the study.
CharacteristicStage A N = 40Stage C N = 40 P
  1. Normal ranges are indicated in parentheses. NS = P > 0·05; *P < 0·05; **P < 0·005; ***P < 0·001.

Gender (Male/Female)25/1525/15NS
Age (years)70 ± 371 ± 3NS
Disease duration (years)6 ± 28 ± 2NS
Body mass index (20·0–25·0 kg/m2)26·4 ± 0·9925·4 ± 1·22NS
Performance Status0·52 ± 0·201·45 ± 0·27 ***
Haemoglobin (115–160 g/l)134 ± 4·0107 ± 6·0 ***
Haematocrit (37–47%)39·9 ± 1·132 ± 1·9 ***
Mean corpuscular volume (80–100 fl)94·5 ± 1·799·9 ± 2·9 **
Mean corpuscular haemoglobin concentration (320–360 g/l)333 ± 20332 ± 4·0NS
Platelet count (150–400 × 109/l)200 ± 17111 ± 19 ***
White blood cell count (4–10 × 109/l)37·0 ± 12·650·2 ± 31·4NS
Lymphocyte count (1-4 × 109/l)30·3 ± 12·145·5 ± 31·3NS
Reticulocytes (80–100 × 109/l)43·0 ± 6·461·4 ± 21·7NS
Thyroid-stimulating hormone (0·2–3·4 miu/l)1·60 ± 0·111·47 ± 0·36NS
Folic Acid (6–39 nmol/l)16·1 ± 1·823·3 ± 6·9NS
Vitamin B12 (142–725 pmol/l)312·9 ± 40·2429·8 ± 78·7NS
Serum iron (10–30 μmol/l)15·0 ± 1·216·0 ± 3·6NS
Transferrin saturation (25–35%)25·0 ± 2·331·5 ± 8·5 ***
Ferritin (10–300 μg/l)117 ± 33431 ± 199 *
Creatinine (60–100 μmol/l)106·6 ± 7·797·1 ± 8·5 *
Aspartate transaminase (0–30 iu/l)23 ± 227 ± 5NS
C-reactive protein (0–6 mg/l)3·0 ± 1·510·0 ± 5·5 *
Albumin (40–48 g/l)43·9 ± 0·740·0 ± 1·9 **
Vitamin C (26·1–84·6 μmol/l)58·7 ± 7·537·7 ± 8·3 ***

To our knowledge, this study is the first to measure the frequency of hypovitaminosis C in CLL patients, and to detect an association between stage C and a lower level of Vitamin C.

Previous studies of vitamin levels reported low levels of vitamin D in CLL. Two groups reported that vitamin D deficiency is associated with inferior time-to-first-treatment and overall survival (Shanafelt et al, 2011; Molica et al, 2012).

Our work highlights the fact that hypovitaminosis C is underestimated in CLL patients. As a known cause of anaemia and fatigue, practitioners should consider screening their patients for vitamin C deficiency. When needed, careful supplementation should be provided together with close surveillance because vitamin C intake may increase markers of haemolysis in patients with sickle cell anaemia or kidney stone formation (Arruda et al, 2013; Thomas et al, 2013). However, the reason for this deficiency remains unclear. Several assumptions can be proposed to explain the low blood levels in CLL: poor nutritional status with low vitamin C intake, reduced vitamin C absorption, or an increased intracellular uptake of vitamin C by tumour cells (Fain, 2004).

These results also raise the question of the possible role of vitamin C in the progression and the treatment of the disease. Our work has some limitations that prevent us from drawing the conclusion that a low vitamin C blood level leads to disease progression. First, dietary differences were not evaluated. Second, we arbitrarily compared two groups of patients at different stages of the disease, which leads to expected differences in performance status, albumin levels, and inflammatory markers, such as ferritin and CRP. Moreover a low vitamin C blood level could result from disease progression. However, a recent in vitro study found that both arsenic trioxide (ATO) and vitamin C mediated cytotoxicity on CLL B lymphocytes and emphasized the ability of vitamin C to increase the efficacy of ATO (Biswas et al, 2010). Vitamin C derivatives have been recently reported as a ‘promising new class of anti-cancer drugs, either alone or in combination with other molecules’, and vitamin C showed efficiency in a phase I clinical trial in biopsy-proven advanced and/or progressing metastatic solid tumours (Bordignon et al, 2013; Stephenson et al, 2013). Thus, we believe that clinical trials and experimental studies should be performed to determine if vitamin C supplementation could indeed maintain stage A CLL quiescence and/or enhance treatment efficacy in patients needing anticancer therapy.


BC, CD, LB, MS, FC, SL, ME, HB, PV, FM, PC and EG performed the research. BC, CD, LB, EG, SL, ME designed the research study. BC, CD, FC, MS, FM, HB, PV, PC and EG analysed the data. BC and EG wrote the paper.

Conflict of interest

The authors have no competing interests.