A high prevalence of malnutrition has been found in patients with cancer, present in 40% to 80% of cases. It has been suggested that malnutrition is one of the main causes of medical complications and even death. Many studies have focused on patients with solid tumors, but leukemia has seldom been investigated. Owing to a higher survival rate and gradually appearing side-effects of malnutrition during growth stages, most studies have been among children rather than adults with leukemia. At present, there is no “gold standard” for the diagnosis or classification of malnutrition.[3,4] Various nutritional evaluation tools such as body mass index (BMI), weight loss, biochemical indexes, and body composition are applied in different studies, yielding inconsistent results.[5–10] This situation is a main factor limiting leukemia-related nutritional research.
Patient-generated subjective global assessment (PG-SGA) is a widely used nutritional screening tool designed especially for patients with cancer.[11,12] It is highly recommended by the American Society for Parenteral and Enteral Nutrition and Anti-Cancer Association of China. PG-SGA consists of 2 parts, one completed by patients and the other completed by medical staff, and includes the indexes such as weight loss history, dietary intake, stress score, and physical examination to qualitatively and quantitatively evaluate nutritional status. A high score indicates a higher risk of malnutrition and critical need of nutrition support. PG-SGA has a high degree of inter-rater reproducibility and high sensitivity and specificity when compared with other validated nutrition assessment tools that have been proved in patients with solid tumor.[13–15] The indexes in PG-SGA are easily collected in patients with leukemia; however, the efficiency of PG-SGA and its relevance with prognosis are unknown.
Our hospital has evaluated nutritional status using PG-SGA in adult patients with newly diagnosed acute leukemia (AL) since 2013. In the current study, we reviewed data of these patients to investigate the feasibility of PG-SGA as a tool to evaluate nutritional status, and to explore the correlation between nutritional status and prognosis.
2 Materials and methods
In this retrospective study, adult patients with newly diagnosed AL in the First Hospital of Jilin University who met the inclusion criteria were enrolled between May 2013 and July 2015. Inclusion criteria were: patients more than 18 years old with a new morphological and immunological diagnosis of acute myeloid leukemia (AML) or acute lymphocytic leukemia (ALL), and with nutritional status evaluated using PG-SGA before drug administration. Exclusion criteria were: patients with a history of myelodysplastic syndrome or myeloproliferative disease, a history of other malignant tumors, those who did not receive treatment after diagnosis, patients with acute promyelocytic leukemia, and those who received allogeneic hematopoietic stem cell transplantation in follow-up treatment. All patients were followed-up until December 31, 2015. All participating patients gave their written informed consent prior to enrollment in the study. This study was approved by the ethics committee of the First Hospital of Jilin University and conducted in accordance with the principles of the Declaration of Helsinki.
2.2 Nutritional data collection
Biochemical indexes such as albumin, C-reactive protein (CRP), transferrin, and triglyceride were detected before therapy using blood tests. Body composition characteristics such as body fat ratio, muscle mass, water fraction, and basal metabolism were measured using a body composition analyzer (InbodyS10, Seoul, Korea), and anthropometric measures such as body weight and nondominant hand grip were measured by trained nurses in our department.
2.3 PG-SGA score classification
PG-SGA was carried out using a questionnaire for the assessment of nutritional status. A doctor and a nurse are required to complete the questionnaire together, with strict quality control. In our study, any disagreement was resolved by a third party. According to PG-SGA nutritional triage recommendations, patients scoring 0 to 1 are considered well nourished; a score of 2 to 3 points indicates suspected malnutrition, 4 to 8 points indicates moderate malnutrition, and 9 points or more indicates severe malnutrition, which requires improved symptom management and/or nutrient intervention options. Patients were divided into a severe malnutrition and a non-severe malnutrition group according to PG-SGA score.
2.4 Clinical data collection and analysis
We collected information of clinical features such as age, sex, risk category, and Karnofsky Performance Status (KPS); chemotherapeutic regimens; complete remission (CR) rate; minimal residual disease (MRD); side-effects of chemotherapy; overall survival (OS), and event-free survival (EFS). MRD was detected by flow cytometry (MACS Quant; Germany) on the 14th day after induction therapy in AML patients and on the 14th and 28th day in ALL patients. Risk category and remission status were based on the 2016 National Comprehensive Cancer Network guideline. ALL patients were divided into high-risk and low-risk groups. AML patients with no response after induction chemotherapy were classified into the high-risk group. To facilitate the data analysis, low- and moderate-risk AML patients were included in the low-risk group.
Treatments for AML patients have been previously described. For ALL patients older than 35 years, chemotherapy followed the 2008 China Acute Lymphocytic Leukemia Group protocol; treatment in the remaining patients followed the 2008 Chinese Children's Leukemia Group–ALL protocol.
Univariate analysis was conducted for age, sex, type of leukemia, risk category, initial leukocyte count, and onset to admission time. Univariate and multivariate analysis was carried out using biochemical indexes, body composition factors, and anthropometric measures between the severe and non-severe malnutrition groups. Induction remission rate, MRD, and survival time were compared between the 2 groups. In both univariate and multivariate analysis, the missing data (Table S3, http://links.lww.com/MD/C83) were deleted because of the small sample size.
2.5 Statistical analysis
Statistical analyses were carried out using SPSS version 21.0 (IBM Corp., Armonk, NY). The χ2, nonparametric, and t-test were used, as appropriate, to make comparisons between groups. General linear regression was applied to find correlations between nutritional status (defined by PG-SGA) and clinical features. OS was measured from diagnosis to the date of death or until the last follow-up if the patient remained alive or dropped out of the study. EFS was measured from diagnosis to the date of induction failure, disease relapse, or death (whichever occurred first), or until the last follow-up if no events occurred. Survival time was analyzed using the Kaplan–Meier log-rank test. Multivariate analysis was carried out by Cox regression. Statistical significance was set at the P < .05 level.
A total of 68 adult patients with newly diagnosed AL met the study criteria, comprising 35 men and 33 women. There were 51 patients with AML and 17 with ALL. Mean age was 43.0 ± 14.8 years (19–83 years). Median follow-up time was 8.25 months (1–30 months) to the end of follow-up; 44 of 68 (64.7%) patients remained alive, 16 (23.5%) died, and 7 (11.8%) dropped out. About 26 patients had events that terminated EFS.
3.1 Correlations between clinical features and PG-SGA score
Average PG-SGA scores were lower in ALL patients than those in AML patients, with statistical significance (4.0 ± 3.2 vs. 6.8 ± 4.0, P = .011). In both AML and ALL patients, average PG-SGA scores were higher in high-risk patients than low-risk patients (AML, 8.8 ± 5.1 vs. 6.2 ± 3.4, P = .012; ALL, 4.4 ± 3.4 vs. 2.0 ± 1.0, P = .043). There were no evident differences in terms of age, sex, initial leukocyte count, and onset to admission time (Table 1).
3.2 Comparison of general characteristics, biochemical indexes, body composition, and anthropometric measures between patients with and without severe malnutrition
According to PG-SGA score, 44 (64.7%) patients were malnourished and 19 (27.9%) were severely malnourished. Compared with those who were not severely malnourished, patients with severe malnutrition had significantly higher average age and CRP levels, as well as lower transferrin and KPS scores. No differences were found for other clinical features, such as sex, comorbidities, and AL subtype (Table 2). In the severe malnutrition group, 10 patients had pneumonia, 4 had gastroenteritis, and 8 had upper respiratory tract infection. In the other group, 8 patients had pneumonia, 2 had gastroenteritis, and 6 had upper respiratory tract infection, with coinfection in some patients. There was no evident difference between the 2 groups. According to multivariate analysis, only CRP was associated with severe malnutrition (Table 3). These results indicated that high CRP level may serve as a predictor of severe malnutrition.
3.3 Comparison of induction remission rate, MRD, and side-effects between groups
No significant difference was found between the 2 nutritional groups for CR rate (12/18 in severe malnutrition patients vs. 34/49 in non-severe malnutrition patients, χ2 = 0.045, P = .831; 1 patient in the severe malnutrition group was not evaluated). However, in AML and ALL patients, distinctions were found in terms of MRD. On the 14th day after chemotherapy in AML patients and the 14th day of chemotherapy in ALL patients, MRD was higher in the severe malnutrition group than in their counterparts in the other group (Table 4).
There was a significantly higher incidence of gastrointestinal side-effects in patients with severe malnutrition (χ2 = 6.058, P = .014). There were no statistically significant differences for other chemotherapy-related side-effects (Table 5).
3.4 Effects of severe malnutrition on survival time
The mean OS of the 68 patients was 22.0 ± 1.7 months; the median OS was not achieved (Fig. S1A, http://links.lww.com/MD/C83). The mean EFS was 17.7 ± 1.8 months and the median EFS was 16.0 ± 2.6 months (Fig. S1B, http://links.lww.com/MD/C83). As shown in Fig. S2A and B, http://links.lww.com/MD/C83 patients with severe malnutrition had significantly worse OS (log-rank = 0.030), and EFS tended to be poor in these patients, with no statistical significance (log-rank = 0.240). According to Cox regression, nutritional status, induction remission status, and risk category were independent factors influencing OS (hazard ratio [HR]: 0.243, 95% confidence interval [CI]: 0.063–0.945, P = .041; Table S1, http://links.lww.com/MD/C83), and patients with severe malnutrition had inferior outcomes. No distinct differences were found in EFS between the 2 nutritional groups (HR: 0.808, 95% CI: 0.338–1.934, P = .663; Table S2, http://links.lww.com/MD/C83); only risk category was associated with EFS.
Diagnosis of malnutrition is a serious problem in patients with tumor including leukemia. Previous studies have showed that the influence of BMI on survival time differs from each other,[5–9] and other indexes such as weight and albumin are also unstable. PG-SGA is highly recommended by nutritionists and can be easily applied in leukemia patients. Thus, we aimed to evaluate the nutritional status of adult patients with AL using PG-SGA and to investigate the influence of nutritional status on prognosis.
In our study, a high prevalence of malnutrition was found in AL patients. A total 64.7% of patients were diagnosed with malnutrition, and there were 27.9% of patients with severe malnutrition who were in critical need of nutritional support, which is slightly more than the proportion of children and young adults, whose nutritional regulation and compensation capacities are stronger.
The clinical features were compared using PG-SGA score. These results indicated that ALL patients had better nutritional status than AML patients. Myeloid hematopoietic cells are abnormally proliferated in AML, which could decrease the levels of neutrophils, erythrocytes, and platelets, thus leading to infection, anemia, and hemorrhage. At the same time, anorexia, gastrointestinal function disorders, and stomatitis—all related to poor appetite—result in less food intake and weight loss, which can be severe. These developments may be the reason why these patients tend to have poor nutritional status. A previous study came to the same conclusion. In high-risk patients, average PG-SGA scores were significantly higher, which indicated that nutritional status was associated with risk category. Several studies among patients with solid cancers have reached the similar conclusion.[22,23] Leukemia cells in high-risk patients tended to be more progressive than those in patients with lower risk, which indicates that progressiveness influenced the function of major organs and might be related to malnutrition and lead to poor outcomes. There was no evidence of differences in other clinical characteristics; however, patients aged more than 60 years and patients with onset to admission time of more than 7 days tended to have higher scores.
It has been reported that PG-SGA scores are influenced by the types and stages of tumor, KPS score, comorbidities, and whether patients have received nutritional support. Based on our data, age, KPS score, CRP level, and transferrin were associated with severe malnutrition in univariate analysis. However, in multivariate analysis, only CRP level was associated with severe malnutrition. CRP is one of the most sensitive and recurrent markers of inflammation. The association between CRP as an indicator of inflammation and nutritional status has been previously reported by others.[20,24–28] High-risk disease may not only increase metabolic demands on the host, leading to malnutrition, but it may also result in increased production and release of proinflammatory cytokines that in turn raise the production of CRP. Thus, elevated CRP levels might be related to both infection and severity of disease. Based on our data, there was an evident difference in CRP and risk category, but not incidence of infection between the 2 groups. Moreover, in patients without pre-existing infection, there was no difference in terms of CRP levels between the 2 groups (data not shown). Therefore, severity of disease might be the main cause of elevated CRP; however, the exact mechanism requires further investigation. No difference was found in other nutritional indexes.
Effects of severe malnutrition evaluated by PG-SGA on efficacy and survival time were also analyzed. No difference was found in CR rate; nevertheless, there was a significantly higher incidence of gastrointestinal side-effects in patients with severe malnutrition, which may be correlated with rapid proliferation and a high need for nutrients in the gastrointestinal mucosa. These results are consistent with those of other studies.[14,30] Though no difference was found in the CR rate, MRD was higher in the severe malnutrition group, which indicates that although severe malnutrition had no significant influence on initial disease relief, there were still differences in deep remission. This might be related to metabolism of chemotherapy drugs.[31,32] The levels of MRD detected on the 28th day of chemotherapy in ALL patients were indistinct between the 2 groups; however, only 2 ALL patients had severe malnutrition. The results clearly require further study with a larger number of patients. To our knowledge, MRD has not been previously analyzed to find a correlation with nutritional status. OS and EFS were compared between the 2 groups; univariate analysis indicated that patients with severe malnutrition had shorter survival time. In addition to nutritional status, the type of AL, age, initial leukocyte count, and risk category were all considered in assessing the influence of nutritional status on survival time. The results indicated that de novo patients who did not have severe malnutrition had superior prognosis than those who were severely malnourished. Because AL subtypes, therapeutic regimen, and CR rate were not different between the 2 groups (Tables 3 and 4), the difference in OS might be correlated with a higher level of MRD and worse risk category in severely malnourished patients; however, no marked difference was found in EFS (Fig. 1A and B). Our results are in accordance with those of previous studies;[6,7,11] but disagree with the findings of others.[5,8] BMI and weight were widely used in studies focusing on the influence of nutritional status on survival time; however, the same conclusions were not reached in these studies, suggesting that these indexes may be unreliable for assessing severe malnutrition.
As far as we are aware, this is the first study to evaluate the correlation between nutritional status and prognosis in adult patients with AL. Owing to the limited sample size, prospective studies with a larger number of patients are clearly required.
In conclusion, malnutrition is commonly seen in leukemia patients and PG-SGA can be used as an efficient tool to evaluate nutritional status. CRP was an independent risk factor of severe malnutrition. Although there was no difference in CR rate between the 2 groups, our study suggests that severely malnourished patients have more side-effects of chemotherapy, higher levels of MRD, and shorter survival time. Thus, for patients with severe malnutrition, nutrient support might be needed before or during chemotherapy. Whether nutrient support can improve survival time and decrease side-effects requires further clinical investigation.
. Isenring E, Bauer J, Capra S. The scored patient-generated subjective global assessment
(PG-SGA) and its association with quality of life in ambulatory patients receiving radiotherapy. Eur J Clin Nutr 2003;57:305–9.
. Van CE, Arends J. The causes and consequences of cancer-associated malnutrition. Eur J Oncol Nurs 2005;9(suppl. 2):S51–63.
. Rogers PC, Melnick SJ, Ladas EJ, et al. Children's Oncology Group (COG) Nutrition Committee. Children's Oncology Group (COG) Nutrition Committee. Pediatr Blood Cancer 2008;50(2 suppl):447–50.
. Brinksma A, Huizinga G, Sulkers E, et al. Malnutrition in childhood cancer patients: a review on its prevalence and possible causes. Crit Rev Oncol Hematol 2012;83:249–75.
. Chen SH, Jaing TH, Hung IJ, et al. High body mass index did not result in poor outcome in Taiwanese children with acute myeloid leukemia: a single institution experience. Int J Hematol 2015;102:48–52.
. Lange BJ, Gerbing RB, Feusner J, et al. Mortality in overweight and underweight children with acute myeloid leukemia. JAMA 2005;293:203–11.
. Butturini AM, Dorey FJ, Lange BJ, et al. Obesity and outcome in pediatric acute lymphoblastic leukemia. J Clin Oncol 2007;25:2063–9.
. Hijiya N, Panetta JC, Zhou Y, et al. Body mass index does not influence pharmacokinetics or outcome of treatment in children with acute lymphoblastic leukemia. Blood 2006;108:3997–4002.
. Inaba H, Surprise HC, Pounds S, et al. Effect of body mass index on the outcome of children with acute myeloid leukemia. Cancer 2012;118:5989–96.
. Li R, Wu J, Ma M, et al. Comparison of PG-SGA, SGA and body-composition measurement in detecting malnutrition among newly diagnosed lung cancer patients in stage IIIB/IV and benign conditions. Med Oncol 2011;28:689–96.
. Tong H, Isenring E, Yates P. The prevalence of nutrition impact symptoms and their relationship to quality of life and clinical outcomes in medical oncology patients. Support Care Cancer 2009;17:83–90.
. Capuano G, Gentile PC, Bianciardi F, et al. Prevalence and influence of malnutrition on quality of life and performance status in patients with locally advanced head and neck cancer before treatment. Support Care Cancer 2010;18:433–7.
. Persson C, Sjödén PO, Glimelius B. The Swedish version of the patient-generated subjective global assessment
of nutritional status
: gastrointestinal vs urological cancers. Clin Nutr 1999;18:71–7.
. Bauer J, Capra S, Ferguson M. Use of the scored Patient-Generated Subjective Global Assessment
(PG-SGA) as a nutrition assessment tool in patients with cancer. Eur J Clin Nutr 2002;56:779–85.
. Desbrow B, Bauer J, Blum C, et al. Assessment of nutritional status
in hemodialysis patients using patient-generated subjective global assessment
. J Ren Nutr 2005;15:211–6.
. Malihi Z, Kandiah M, Chan YM, et al. The effect of dietary intake changes on nutritional status
in acute leukeamia patients after first induction chemotherapy. Eur J Cancer Care (Engl) 2015;24:542–52.
. Su L, Gao S, Liu X, et al. CEBPA mutations in patients with de novo acute myeloid leukemia: data analysis in a Chinese population. Onco Targets Ther 2016;9:3399–403.
. Hu Y, He H, Lu J, et al. E2A-PBX1 exhibited a promising prognosis in pediatric acute lymphoblastic leukemia treated with the CCLG-ALL2008 protocol. Onco Targets Ther 2016;9:7219–25.
. Lou Y, Ma Y, Li C, et al. Efficacy and prognostic factors of imatinib plus CALLG2008 protocol in adult patients
with newly diagnosed Philadelphia chromosome-positive acute lymphoblastic leukemia. Front Med 2017;11:229–38.
. Owens JL, Hanson SJ, McArthur JA, et al. The need for evidence based nutritional guidelines for pediatric acute lymphoblastic leukemia patients: acute and long-term following treatment. Nutrients 2013;5:4333–46.
. Esfahani A, Ghoreishi Z, Abedi Miran M, et al. Nutritional assessment of patients with acute leukemia
during induction chemotherapy: association with hospital outcomes. Leuk Lymphoma 2014;55:1743–50.
. Carvalho CS, Souza DS, Lopes JR, et al. Relationship between patient-generated subjective global assessment
and survival in patients in palliative care. Ann Palliat Med 2017;6(suppl 1):S4–12.
. Tan CS, Read JA, Phan VH, et al. The relationship between nutritional status
, inflammatory markers and survival in patients with advanced cancer: a prospective cohort study. Support Care Cancer 2015;23:385–91.
. Read JA, Choy ST, Beale PJ, et al. Evaluation of nutritional and inflammatory status of advanced colorectal cancer patients and its correlation with survival. Nutr Cancer 2006;55:78–85.
. Vigano AL, di Tomasso J, Kilgour RD, et al. The abridged patient-generated subjective global assessment
is a useful tool for early detection and characterization of cancer cachexia. J Acad Nutr Diet 2014;114:1088–98.
. McMillan DC. Systemic inflammation, nutritional status
and survival in patients with cancer. Curr Opin Clin Nutr Metab Care 2009;12:223–6.
. Slaviero KA, Read JA, Clarke SJ, et al. Baseline nutritional assessment in advanced cancer patients receiving palliative chemotherapy. Nutr Cancer 2003;46:148–57.
. Arrieta O, Michel Ortega RM, Villanueva-Rodriguez G, et al. Association of nutritional status
and serum albumin levels with development of toxicity in patients with advanced non-small cell lung cancer treated with paclitaxel-cisplatin chemotherapy: a prospective study. BMC Cancer 2010;10:50.
. Deans DA, Tan BH, Wigmore SJ, et al. The influence of systemic inflammation, dietary intake and stage of disease on rate of weight loss in patients with gastro-oesophageal cancer. Br J Cancer 2009;100:63–9.
. Martin L, Watanabe S, Fainsinger R, et al. Prognostic factors in patients with advanced cancer: use of the patient-generated subjective global assessment
in survival prediction. J Clin Oncol 2010;28:4376–83.
. Walter-Sack I, Klotz U. Influence of diet and nutritional status
on drug metabolism. Clin Pharmacokinet 1996;31:47–64.
. Prado CM, Maia YL, Ormsbee M, et al. Assessment of nutritional status
in cancer–the relationship between body composition and pharmacokinetics. Anticancer Agents Med Chem 2013;13:1197–203.