Annals of Leukemia Research (ALR)
ISSN: 2688-9587 | Vol. 6, Issue 1, 2025, pp. 29-34

Nutritional Status and Treatment Tolerance in Children Undergoing Leukemia Induction Therapy

Egle Petraitiene1, Federica Rinaldi2, Michal Wozniak3

1Department of Bioinformatics, Vilnius University, Vilnius, Lithuania
2Department of Haematology, University of Turin, Turin, Italy
3Department of Cardio-Oncology, Medical University of Warsaw, Warsaw, Poland
DOI: 10.5281/zenodo.22238196 | Published: 2025-12-22 | Article page and PDF

Abstract

Nutritional status at diagnosis predicts survival in childhood leukaemia, and the literature disagrees about how much. A large part of that disagreement is not biological: body mass index performs poorly in a child whose weight includes tumour bulk and steroid-driven fluid, and arm measurements detect what it misses. Objectives: To review the published evidence on nutritional status in children undergoing induction therapy for acute leukaemia: how often malnutrition is present, how it is measured, what it predicts for treatment tolerance and survival, and how status changes across induction itself. Material and Methods: Narrative review of observational cohorts, retrospective analyses, systematic reviews and nutritional assessment guidance indexed in PubMed and the major haematology, paediatric and nutrition journals. Studies were eligible if they reported the prevalence, measurement or clinical

consequences of nutritional status in a defined paediatric acute leukaemia population. No new patient data were generated and no patients were recruited; every estimate quoted is one published by the original investigators. Results: Reported malnutrition at diagnosis ranges from 0 to 70 per cent across settings, with around 50 per cent in low and middle-income countries, 58.4 per

cent in a Mexican cohort and 45.3 per cent classified underweight in an Indian cohort. In the Mexican series, survival differed by body mass index category at 76 against 63 per cent (p = 0.049), while relapse differed more sharply by triceps skinfold, at 41 against 12 per cent below and above minus two standard deviations (p = 0.007). Stunting carried a hazard ratio for mortality of 6.214 (95% CI 1.372-28.154) and low triceps skinfold 2.91 (95% CI 1.27-6.68), while higher mid-upper arm circumference was protective at 0.85 (95% CI 0.73-1.00). Across induction in a European cohort of 50 children, the proportion with an abnormal body mass index rose from 6 to 20 per cent. Conclusion: Arm anthropometry outperforms weight-based measures in this population and produces the stronger associations wherever both have been reported. Nutritional status also deteriorates measurably during induction, so a single assessment at diagnosis describes a moving target. Abbreviations: ALL – Acute Lymphoblastic Leukaemia, AML – Acute Myeloid Leukaemia, BMI – Body Mass Index, MUAC – Mid-Upper Arm Circumference, TSF – Triceps Skinfold Thickness, SD – Standard Deviation, HR – Hazard Ratio, LMIC – Low and Middle-Income Country, HIC – High-Income Country, WHO – World Health Organization, MRD – Minimal Residual Disease, FM – Fat Mass, FFM – Fat-Free Mass.

Keywords: Paediatric Leukaemia; Nutritional Status; Induction Therapy; Anthropometry; Treatment Tolerance

Introduction

The proposed mechanisms are plausible and multiple. Low body mass and lean tissue impair immune function and increase susceptibility to infection; reduced capacity to metabolise cytotoxic drugs raises the risk of organ toxicity; and recovery is prolonged through extended neutropenia and impaired tissue repair [9,10]. Increased body fat has been linked to the same complications, so fat may play a dual role [9,11].

This review sets out how often malnutrition is present at diagnosis, how it is measured and why that matters, what it predicts for tolerance and survival, and how status changes during induction itself.

Material and Methods

Search strategy: PubMed and the archives of the major haematology, paediatric oncology and clinical nutrition journals were searched for studies of nutritional status in childhood leukaemia. Search terms combined acute leukaemia and paediatric with nutritional status, malnutrition, undernutrition, body mass index, anthropometry, mid-upper arm circumference, treatment tolerance and induction. Reference lists of retrieved cohorts and systematic reviews were screened by hand.

Eligibility: Studies were included if they reported the prevalence, assessment method or clinical consequences of nutritional status in a defined paediatric acute leukaemia population, or evaluated a nutritional intervention in that population. Studies in mixed paediatric cancer cohorts are cited where they address measurement methodology and are identified as such.

Data extraction and handling: Design, cohort size, setting, assessment method, threshold applied and reported outcome estimates were extracted as published. The assessment method is stated alongside every prevalence and outcome figure, because weight-based and arm-based measures classify different children as malnourished and this accounts for much of the variation between studies. No pooling or re-analysis was performed and no patient-level data were obtained. Figures 1 to 4 and Tables 1 to 3 reproduce published values only. This is a narrative synthesis and was not registered as a systematic review.

Results

How common malnutrition is, and why estimates vary:

Undernutrition among children and adolescents receiving cancer treatment has been reported across a range from 0 to 70 per cent [12], which is wide enough to require explanation. Around 50 per cent of children with acute lymphoblastic leukaemia in low and middle-income countries are malnourished at diagnosis [4,13], and a Mexican retrospective cohort reported 58.4 per cent (Figure 1) [9].

A prospective Indian study of 64 children and adolescents aged 2 to 20 classified 45.31 per cent as underweight by body mass index, 51.56 per cent as healthy weight and 3.13 per cent as overweight [12]. In high-income settings the balance inverts, with a significant proportion of children with leukaemia categorised as overweight or obese rather than underweight [7,8,14].

Part of the variation is genuinely geographic and part is methodological. Different studies use body mass index, weight for age, arm anthropometry or combinations, applied against different reference standards and thresholds, and these do not identify the same children [3,15,16].

The direction of the association also differs by income setting. A meta-analysis of body mass index and survival in paediatric leukaemia found poorer outcome at both extremes [7], and mortality was elevated in overweight and underweight children with myeloid leukaemia in a large cooperative group analysis [8]. Obesity has been examined separately as a prognostic factor in lymphoblastic disease

Figure 1: Reported prevalence of malnutrition at diagnosis across settings [4,9,12,13]
Figure 1: Reported prevalence of malnutrition at diagnosis across settings [4,9,12,13]

The measurement problem: Arm anthropometry provides more accurate information on nutritional status than measures based on body weight in children with cancer [3]. The reason is specific to this population: weight at diagnosis includes tumour bulk, organomegaly and often effusions, and during induction it includes corticosteroid-driven fluid retention and appetite stimulation, none of which reflects nutritional reserve [3,17].

Body composition rather than weight is where the clinically relevant changes appear, specifically the distribution and proportion of fat mass and fat-free mass [4,18]. Mid-upper arm circumference and triceps skinfold thickness sample peripheral soft tissue directly and are less distorted by the disease and its treatment [3,9].

The same cohort can therefore give different answers depending on the instrument, and the Mexican study reported both. Survival differed by body mass index category at 76 against 63 per cent, reaching significance only marginally (p = 0.049) (Figure 2) [9]. Relapse differed by triceps skinfold at 41 per cent below minus two standard deviations against 12 per cent above, with p = 0.007 (Figure 3) [9].

Body composition studies support the same conclusion from a different direction. Children with cancer differ from healthy peers in the proportion of fat and fat-free mass at a given weight [18], so two children with identical body mass index can differ substantially in lean tissue. Guidance on assessment in this population accordingly recommends measures that sample tissue directly [15,23].

Figure 2: Survival classified by body mass index in a retrospective cohort [9]
Figure 2: Survival classified by body mass index in a retrospective cohort [9]
Figure 3: Relapse in the same cohort classified by triceps skinfold thickness [9]
Figure 3: Relapse in the same cohort classified by triceps skinfold thickness [9]

Principal reported

Study n Setting finding

Arm anthropometry Barr et al. 2016 [3] Review LMIC focus more accurate than weight-based measures Assessment tools and Rogers et al. [15] Guidance International screening recommendations 45.3% underweight; Sharma et al. 2024 India, 64 undernutrition worsens [12] prospective across induction Abnormal body mass Petraitiene et al. 2023 50 Lithuania index rose from 6% to [17] 20% Arm measures Mexican cohort 2025 Retrospec Mexico outperformed body [9] tive mass index Nutritional Systematic review Review International interventions during 2024 [4] treatment assessed Table 1: Principal studies reviewed, with setting and the finding as reported by the original investigators What each measure predicts: In multivariable analysis from the same Mexican cohort, stunting carried a hazard ratio for mortality of 6.214 (95% CI 1.372 to 28.154) and triceps skinfold below minus two standard deviations a hazard ratio of 2.91 (95% CI 1.27 to 6.68). Higher mid-upper arm circumference z-score was associated with reduced mortality risk, at 0.85 (95% CI 0.73 to 1.00) [9].

The confidence interval around the stunting estimate is very wide, spanning from 1.4 to 28, which reflects small numbers in that subgroup and should temper any strong reading of the point estimate [9]. The arm measures carry tighter intervals and are the more dependable findings in that analysis.

Malnutrition at diagnosis or during treatment has been linked more broadly to higher infection risk, poorer treatment tolerance and adherence, lower survival and death [4,10,19]. Proposed mediating changes include inflammation, increased energy expenditure, inadequate or excessive caloric intake and altered metabolic pathways [4,18,20].

Status deteriorates during induction: A European observational cohort followed 50 consecutive children with acute lymphoblastic leukaemia through induction using body mass index z-scores against World Health Organization growth standards. The number with an abnormal index rose from 3 (6 per cent) at diagnosis to 10 (20 per cent) at the end of induction, comprising a rise from 2 to 6 in the overweight or obese category and from 1 to 4 in the underweight category (Figure 4) [17].

At the end of induction, all children categorised as overweight or obese were aged between 0 and 5 years [17], which points toward corticosteroid effect in the youngest patients rather than a change in habitual intake. The prospective Indian study found a parallel trend of worsening undernutrition in both underweight and healthy weight groups across the four weeks of induction [12].

Deterioration in both directions during a four-week block has a practical implication. A single assessment at diagnosis, which is what most cohorts report, describes a status that will have changed by the time treatment decisions dependent on it are taken [12,17].

Figure 4: Change in nutritional status across induction in 50 children [17]
Figure 4: Change in nutritional status across induction in 50 children [17]

Performance in this

Measure What it captures

population

Total weight for Distorted by tumour bulk Body mass index height and steroid fluid [3,17] Same distortion; no height Weight for age Total weight adjustment [3] Mid-upper arm Peripheral muscle Protective association, HR circumference and fat 0.85 [3,9] Subcutaneous fat Strongest relapse Triceps skinfold store association, p = 0.007 [9] Stunting (height for Chronic Large effect, very wide age) undernutrition interval [9] Table 2: Assessment measures and how each behaves in children receiving leukaemia therapy What intervention achieves: Nutritional status is potentially modifiable, which distinguishes it from most prognostic factors in this disease, and interventions have been reviewed systematically [4]. Relationships between status and outcome are described as subject to change with nutritional intervention [3], though demonstrating that a specific intervention alters survival is a different and harder claim.

Reduced oral intake during induction has identifiable causes: nausea, vomiting, diarrhoea, mucositis and renal losses all follow from the regimen itself [4,20,21]. These are treatable, and addressing them is supportive care rather than nutritional therapy, which may be why intervention trials have been difficult to design cleanly [4,15]. Weight extremes over time rather than at a point: Analyses using duration of weight extremes rather than a single classification have reported effects on both survival and toxicity, which is a different and arguably better-specified exposure than status at diagnosis [26]. Being underweight and losing weight during therapy each carried adverse weight in a large European cohort [27].

Prospective observation across whole treatment courses shows status changing repeatedly rather than settling [25], and multicentre work following patients from diagnosis through therapy found malnutrition appearing in children who began adequately nourished [14]. Narrative and systematic reviews of assessment have set out how these trajectories should be captured [22,23].

The infection link: The mechanism most often invoked is impaired immunity, and the clinical consequence has been measured. Malnutrition in childhood cancer has been associated specifically with infection and mortality [24], and in a Nicaraguan cohort with treatment-related morbidity as well as survival [28].

That places nutritional status alongside the other determinants of infectious risk during induction, where antibacterial prophylaxis and supportive care standards have their own evidence base [29,30]. Whether nutritional intervention reduces infection independently of those measures has not been established, and separating the two would require a trial designed for the purpose [4,24].

Reported

Exposure definition Source association

Retrospective cohorts Status at diagnosis Survival and relapse [9,14] Duration of weight Trial cohort analysis Survival and toxicity extremes [26] Weight loss during Reduced survival European cohort [27] therapy Change across Not yet linked to Small prospective induction outcome cohorts [12,17]

Table 3: How the exposure has been defined across studies and what each definition predicts
Table 3: How the exposure has been defined across studies and what each definition predicts

The association between nutritional status and outcome in childhood leukaemia is real but reported inconsistently, and the inconsistency has a specific and largely methodological source. Weight-based measures perform poorly here because weight at diagnosis contains tumour, and weight during induction contains corticosteroid-driven fluid [3,17]. Neither reflects nutritional reserve.

Where both instruments have been applied to one cohort, the arm measures produced the stronger and more precise associations: relapse at 41 against 12 per cent by triceps skinfold with p = 0.007, against survival at 76 versus 63 per cent by body mass index with p = 0.049 [9]. That is the clearest available evidence that the choice of instrument, not the underlying biology, drives much of the disagreement in this field.

Why a single assessment is insufficient

Nutritional status is not a baseline characteristic in this setting; it is a variable that moves under treatment. Abnormal body mass index tripled across a single induction block, in both directions [17], and undernutrition worsened across four weeks in a separate prospective cohort including children who began at healthy weight [12].

Studies reporting status only at diagnosis are therefore relating outcome to a measurement that no longer describes the patient by the time complications occur. Serial assessment through induction would be straightforward with arm anthropometry, which requires a tape measure and calipers rather than any equipment a ward would lack [3,15].

The two directions of malnutrition

Undernutrition and overweight are grouped under one definition and are often analysed as one exposure, but their mechanisms differ [5,7]. Undernutrition plausibly impairs immune function and drug metabolism [9,10]; overweight has been linked to altered pharmacokinetics and to the same complications by a different route, so body fat may act in both directions [9,11].

Analysing them together as abnormal status risks cancelling opposing effects. The finding that all overweight children at the end of induction were aged under five [17] further suggests that the overweight category in this setting partly reflects treatment rather than habitual nutrition, which is a different exposure again.

Study limitations

The limitations belong to the underlying literature. Cohorts are small, with the studies reporting serial change enrolling 50 and 64 children respectively [12,17]. Assessment methods, reference standards and thresholds differ between studies and classify different children as malnourished, which accounts for much of the reported prevalence range [3,15,16]. Confidence intervals around several effect estimates are very wide, notably for stunting [9]. Nutritional status correlates with socioeconomic circumstance, healthcare access and treatment abandonment, so residual confounding is substantial in observational designs [3,13]. And most analyses relate outcome to status at diagnosis alone, despite evidence that status changes materially within weeks [12,17].

Two gaps deserve naming. No trial has tested whether correcting nutritional status during induction alters treatment tolerance or survival, so the modifiability that makes this factor attractive remains an inference [3,4]. And no study reviewed here reports outcomes stratified by both direction of malnutrition and assessment method, which is what would separate the measurement problem from the biological one [9,15].

Conclusion

Malnutrition is present at diagnosis in around half of children with acute leukaemia in low and middle-income settings, with reported estimates spanning 0 to 70 per cent across the literature. It predicts survival, relapse and treatment tolerance, but the strength of that prediction depends heavily on how status is measured. In the one cohort reporting both, triceps skinfold separated relapse at 41 against 12 per cent while body mass index separated survival at 76 against 63 per cent only marginally, and mid-upper arm circumference was independently protective. Arm anthropometry outperforms weight-based assessment because weight in this population contains tumour and corticosteroid-driven fluid rather than nutritional reserve. Status also worsens measurably within a single induction block, in both directions, so assessment at diagnosis alone describes a patient who has already changed.

Two changes follow that need no new evidence. Reporting mid-upper arm circumference and triceps skinfold alongside body mass index would let future analyses be compared, which at present they largely cannot be [3,15,23]. And repeating the assessment at the end of induction rather than at diagnosis alone would capture a change that occurs in one child in five [17].

The question that does need a trial is whether correcting status alters outcome. Nutritional status is among the few prognostic factors in this disease that is modifiable, which is what makes it interesting, and that modifiability remains an inference rather than a demonstrated result [3,4].

Acknowledgements

The authors thank the paediatric oncology units and cohort investigators whose published data form the substance of this article.

Funding

No dedicated funding from public, commercial or charitable sources supported this review.

Conflict of Interest

The authors declare no competing interests.

Ethics Statement

This article reviews previously published studies only. No new patient data were collected and no participants were recruited, so ethical approval was not required.

References

  1. Pui, CH., Yang, JJ., Hunger, SP., Pieters, R., Schrappe, M., Biondi, A., et al. (2015) Childhood acute lymphoblastic leukemia: progress through collaboration. J Clin Oncol, 33(27): 2938-2948.
  2. Hunger, SP., Mullighan, CG. (2015) Acute lymphoblastic leukemia in children. N Engl J Med, 373(16): 1541-1552.
  3. Barr, RD., Gomez-Almaguer, D., Jaime-Perez, JC., Ruiz-Arguelles, GJ. (2016) Importance of nutrition in the treatment of leukemia in children and adolescents. Arch Med Res, 47(8): 585-592.
  4. Gonzalez-Torres, LA., Rodriguez-Ramirez, S., Perez-Cruz, E., Serrano-Sanchez, MF. (2024) Nutritional interventions in children with acute lymphoblastic leukemia undergoing antineoplastic treatment: a systematic review. Clin Nutr ESPEN, 62: 291-300.
  5. World Health Organization. (2021) Malnutrition: key facts. Geneva: World Health Organization.
  6. de Onis, M., Onyango, AW., Borghi, E., Siyam, A., Nishida, C., Siekmann, J. (2007) Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ, 85(9): 660-667.
  7. Orgel, E., Genkinger, JM., Aggarwal, D., Sung, L., Nieder, M., Ladas, EJ. (2016) Association of body mass index and survival in pediatric leukemia: a meta-analysis. Am J Clin Nutr, 103(3): 808-817.
  8. Lange, BJ., Gerbing, RB., Feusner, J., Skolnik, J., Sacks, N., Smith, FO., Alonzo, TA. (2005) Mortality in overweight and underweight children with acute myeloid leukemia. JAMA, 293(2): 203-211.
  9. Hernandez-Valdivia, E., Zamora-Nava, LE., Barr, RD., Mejia-Arangure, JM., Nunez-Enriquez, JC., et al. (2025) Nutritional status at diagnosis and its relationship with survival and relapse in Mexican children with acute lymphoblastic leukemia: a retrospective study. BMC Cancer, 25: 318.
  10. Sala, A., Pencharz, P., Barr, RD. (2004) Children, cancer, and nutrition: a dynamic triangle in review. Cancer, 100(4): 677-687.
  11. Butturini, AM., Dorey, FJ., Lange, BJ., Henry, DW., Gaynon, PS., Fu, C., et al. (2007) Obesity and outcome in pediatric acute lymphoblastic leukemia. J Clin Oncol, 25(15): 2063-2069.
  12. Sharma, S., Bakhshi, S., Pandey, RM., Kumar, R., Dhawan, D., Gupta, YK. (2024) Evaluation of nutritional status during induction chemotherapy in patients with acute leukemia. Cureus, 16(11): e73892.
  13. Israels, T., Renner, L., Hendricks, M., Hesseling, P., Howard, S., Molyneux, E. (2013) SIOP PODC: recommendations for supportive care of children with cancer in a low-income setting. Pediatr Blood Cancer, 60(6): 899-904.
  14. Zimmermann, K., Ammann, RA., Kuehni, CE., De Geest, S., Cignacco, E. (2013) Malnutrition in pediatric patients with cancer at diagnosis and throughout therapy: a multicenter cohort study. Pediatr Blood Cancer, 60(4): 642-649.
  15. Rogers, PC., Barr, RD. (2020) The relevance of nutrition to pediatric oncology: a cancer control perspective. Pediatr Blood Cancer, 67(Suppl
  16. : e28213.
  17. Joffe, L., Dwyer, S., Glade Bender, JL., Frazier, AL., Ladas, EJ. (2019) Nutritional status and clinical outcomes in pediatric patients with solid tumors: a systematic review of the literature. Semin Oncol, 46(1): 48-56.
  18. Petraitiene, S., Alasevicius, T., Staneviciene, I., Vaidotaite, G., Rageliene, L., et al. (2023) Changes in nutritional status during induction phase and their association with fever and minimal residual disease in paediatric acute lymphoblastic leukaemia. Medicina (Kaunas), 59(6): 1008.
  19. Murphy, AJ., White, M., Davies, PS. (2010) Body composition of children with cancer. Am J Clin Nutr, 92(1): 55-60.
  20. Antillon, F., Rossi, E., Molina, AL., Sala, A., Pencharz, P., Valsecchi, MG., Barr, R. (2013) Nutritional status of children during treatment for acute lymphoblastic leukemia in Guatemala. Pediatr Blood Cancer, 60(6): 911-915.
  21. Ladas, EJ., Sacks, N., Meacham, L., Henry, D., Enriquez, L., Lowry, G., et al. (2005) A multidisciplinary review of nutrition considerations in the pediatric oncology population. Nutr Clin Pract, 20(4): 377-393.
  22. Sacks, N., Hwang, WT., Lange, BJ., Tan, KS., Sandler, ES., Rogers, PC., et al. (2014) Proactive enteral tube feeding in pediatric patients undergoing chemotherapy. Pediatr Blood Cancer, 61(2): 281-285.
  23. Barr, RD., Stevens, MCG. (2020) The influence of nutrition on clinical outcomes in children with cancer. Pediatr Blood Cancer, 67(Suppl 3): e28117.
  24. Viani, K., Trehan, A., Manzoli, B., Schoeman, J. (2020) Assessment of nutritional status in children with cancer: a narrative review. Pediatr Blood Cancer, 67(Suppl 3): e28211.
  25. Loeffen, EAH., Brinksma, A., Miedema, KGE., de Bock, GH., Tissing, WJE. (2015) Clinical implications of malnutrition in childhood cancer patients: infections and mortality. Support Care Cancer, 23(1): 143-150.
  26. Brinksma, A., Roodbol, PF., Sulkers, E., Kamps, WA., de Bont, ESJM., Boot, AM., et al. (2015) Changes in nutritional status in childhood cancer patients: a prospective cohort study. Clin Nutr, 34(1): 66-73.
  27. Orgel, E., Sposto, R., Malvar, J., Seibel, NL., Ladas, E., Gaynon, PS., Freyer, DR. (2014) Impact on survival and toxicity by duration of weight extremes during treatment for pediatric acute lymphoblastic leukemia. J Clin Oncol, 32(13): 1331-1337.
  28. den Hoed, MAH., Pluijm, SMF., de Groot-Kruseman, HA., te Winkel, ML., Fiocco, M., van den Akker, ELT., et al. (2015) The negative impact of being underweight and weight loss on survival of children with acute lymphoblastic leukemia. Haematologica, 100(1): 62-69.
  29. Pribnow, AK., Ortiz, R., Baez, LF., Mendieta, L., Luna-Fineman, S. (2017) Effects of malnutrition on treatment-related morbidity and survival of children with cancer in Nicaragua. Pediatr Blood Cancer, 64(11): e26590.
  30. Lehrnbecher, T., Fisher, BT., Phillips, B., Alexander, S., Ammann, RA., Beauchemin, M., et al. (2020) Clinical practice guideline for antibacterial prophylaxis in pediatric patients with cancer. Clin Infect Dis, 71(1): 226-236.
  31. Bochennek, K., Hassler, A., Perner, C., Gilfert, J., Schoning, S., et al. (2016) Infectious complications in children with acute myeloid leukemia: decreased mortality in multicenter trial AML-BFM 2004. Blood Cancer J, 6: e382.