2.1. Study design

Review of the scientific literature.

2.2. Eligibility criteria

2.3. Search strategy

We conducted a systematic literature search in the main electronic databases without any restriction of language or publication date, from 1966 until August 2016, limiting to studies that evaluated or described the efficacy and safety of medical procedures used for the management and the prevention of symptoms induced by the treatment of cancer in children.

The main sources of information were

Search terms: interventions, medical interventions, clinical interventions, prevention, treatment, symptoms, symptom relief, induced symptoms, uncomfortable symptoms, cancer, neoplasm, oncological, children, childhood, pediatrics, infants, receiving, undergoing chemotherapy, chemotherapy-induced, radiotherapy, efficacy, safety, adverse effects, side effects, harm associated, systematic review, meta-analysis.

MEDLINE searches were conducted using MeSH terms and keywords. In addition, we included methodological filters by type of article: systematic review, meta-analysis, studies in humans (indexed and with abstract).

2.4. Data collection and extraction

Two reviewers separately assessed the eligibility of the studies for inclusion. Relevant studies were retrieved, and data about the methodological aspects, participants, interventions and outcome variables were extracted. In both phases (selection of the studies and extraction of data), disagreements were resolved by consensus. In the case of a persisting disagreement, a third reviewer was consulted.

Data were registered in tables previously designed by a reviewer using Excel software and double-checked by a second reviewer. The collection of data was tested with a sample of articles.

The following information extracted from the identified studies was included:

2.5. Analysis plan

A qualitative synthesis of the results was presented as evidence tables. Results were represented as numeric values, averages, percentages and according to the outcome measures expressed in the studies. A quantitative synthesis of the results was presented accordingly to as they were analyzed and expressed in the reviews. Secondary studies were identified and included, and the studies examined different types of interventions; hence, no data analysis was conducted. Results are presented in a narrative form.

3.1. Description of the studies

Eight Cochrane systematic reviews that evaluated the efficacy and safety of medical procedures used for the management and prevention of the symptoms induced by the treatment of children with cancer were included.^{15,16,17,18,19,20,21,22}

From the eight reviews included, three systematic reviews were performed in the United Kingdom; the rest were carried out in China, Brazil, Israel, the U. S. and the Netherlands. The reviews included experimental studies, which considered randomized controlled clinical trials (RCT) and controlled clinical trials (CCT) with a parallel design, cross-over trials, and quasi-randomised trials. Four reviews included children and young people;^18,20,21,22 the other four included children and adults.^15,16,17,19 All the participants were patients diagnosed with some cancer about to receive or had already received chemotherapy, radiotherapy, hematopoietic stem cell transplantation or targeted therapy. Three reviews evaluated the efficacy and safety of colony-stimulating factors (CSFs) for preventing or treating infections and febrile neutropenia induced by treatment in patients with cancer.^16,17,20 Two reviews evaluated medical interventions for the prevention of cardiotoxicity induced by the treatment with anthracycline,¹⁵ and the prevention of platinum-induced hearing loss in children with cancer.²¹ A review assessed the effects of oral cryotherapy for the prevention of oral mucositis in patients receiving anticancer treatment.¹⁹ Another review evaluated antiemetic drugs for prevention and treatment of chemotherapy-induced nausea and vomiting in children.¹⁸ The last review determined the effects of nutritional support in children and young people with cancer undergoing chemotherapy.²²

The details of the results of each study are presented in Tables 2 and 3, and key results are shown in Table 4.

Gurion et al. (2011) evaluated the efficacy and safety of the colony-stimulating factor (CSF) for the prevention and treatment of infectious complications in patients with acute myeloid leukemia (AML).¹⁶ Nineteen RCTs with 5,256 participants aged 0 to 88 years were included. Colony-stimulating factors used were granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage-colony-stimulating factor (GM-CSF), administered after chemotherapy induction, consolidation or salvage treatment, and after hematopoietic stem cell transplantation. Placebo or no treatment groups were compared. Most clinical trials included patients > 15 years of age (12 trials, 63%). One test was performed in children aged 0 to 18 of age. Seventeen trials included patients subjected to induction chemotherapy (89%), one trial included patients subjected to consolidation chemotherapy and one trial included patients subjected to salvage chemotherapy. Chemotherapy protocols used in the trials were heterogeneous. They consisted of different combinations of anthracyclines and cytarabine with or without etoposide. According to the results, 14 trials with 4,119 participants reported all-cause mortality. Meta-analysis of data from these studies indicated that there was no difference in all-cause mortality among chemotherapy and CSFs groups compared with chemotherapy alone, with a risk ratio (RR) of 1.01 (95% confidence interval (95% CI) 0.98 to 1.05). For overall survival, a meta-analysis of 11 trials involving 3,335 participants reported no significant difference between patients treated with chemotherapy and CSFs compared with chemotherapy alone, with a hazard ratio (HR) of 1.00 (95% CI 0.93 to 1.08). The use of CSFs did not decrease the occurrence of febrile neutropenia episodes (RR 0.98; 95% CI 0.94 to 1.03), bacteremia (RR 0.96; 95% CI 0.82 to 1.12), or the incidence of invasive fungal infections (RR 1.40; 95% CI 0.90 to 2.19). The results also reported that CSFs significantly reduced the duration of neutropenia and hospital stay. Only two trials reported adverse event data: a study reported 11 cases of events in the CSFs group, and the other study reported 25 cases in the CSFs group compared with nine cases in the control group. The review concluded that CSFs should not be given routinely to patients with AML after chemotherapy since they do not improve survival or the prevalence of infections such as bacteremia and invasive mycosis. However, they do not negatively affect hematological outcomes, complete remission, relapse rates or disease-free survival.

Mhaskar et al. (2014) evaluated the safety and efficacy of the addition of granulocyte colony-stimulating factor (G-CSF) or granulocyte-macrophage-colony-stimulating factor (GM-CSF) to the standard antibiotic treatment for febrile neutropenia induced by chemotherapy in people diagnosed with cancer.¹⁷ In this review, 14 RCTs with 1,553 participants were included. From these trials, one included children and adults with hematological malignancies; three included children with mixed tumors, and the rest of the trials included adult participants presenting both mixed tumors and hematological malignancies (10 trials, 71%). The studies compared the use of a CSF plus antibiotics versus antibiotics alone. Seven studies evaluated the effects of G-CSF, six studies evaluated the effects of GM-CSF and one study evaluated G-CSF, GM-CSF, and placebo. According to the outcomes, a meta-analysis that grouped 13 trials (1,335 participants) showed that CSFs do not diminish the overall mortality (HR 0.74, 95% CI 0.47 to 1.16; p = 0.19), or the mortality associated with infection (10 trials with 897 participants, HR 0.75, 95% CI 0.47 to 1.20; p = 0.23). However, a meta-analysis of eight trials (1,221 participants) showed that CSF plus antibiotics reduced the number of patients hospitalized for more than ten days (RR 0.65, 95% CI 0.44 to 0.95; p = 0.03). Similarly, the analysis of five trials (794 participants) for neutrophil recovery time showed a statistically significant effect of the CSF plus antibiotics compared to antibiotics alone (RR 0.52, 95% CI 0.34 to 0.81; p = 0.004). For the duration of the neutropenia grade IV, time recovery from fever and the duration of antibiotics use, data analysis showed a statistically significant effect in favor of the CSF plus antibiotics compared to antibiotics alone. However, patients that received CSF plus antibiotics presented a higher incidence of symptoms such as bone and joint pain compared with patients treated with antibiotics alone. The review concludes that the use of CSF plus antibiotics in individuals with chemotherapy-induced febrile neutropenia has no effect on overall mortality, but reduces the time of hospital stay and improves neutrophil recovery.

Sasse et al. (2005)²⁰ conducted a systematic review to evaluate the safety and effectiveness of the addition of CSF for the prevention of febrile neutropenia induced by myelosuppressive chemotherapy in children with acute lymphoblastic leukemia (ALL). The review included six RCTs with 332 participants: five trials examined the effects of G-CSF, and one trial analyzed the effects of GM-CSF. The majority of the studies used CSF as primary prophylaxis; only one study used CSF as secondary prophylaxis. These results reported that only two studies where patients received 451 cycles of treatment presented data on febrile neutropenia episodes, where the CSF group (226 cycles) reported 68 episodes compared with the control group (228 cycles) that presented 112 episodes (RR -0.47, 95% CI -0.77 to -0.16). These results showed a risk reduction of 47% of presenting febrile neutropenia with the use of CSF. Only two studies reported deaths: one by septic shock and three more by treatment-related toxicity. Three studies (134 participants) reported data on the duration of neutropenia. The combined data from the studies suggested that patients who received CSF had a shorter duration of neutropenia compared with patients who did not receive CSF (weighted mean difference (WMD) = - 3.44, 95% CI - 4.76 to - 2.12; p < 0.00001). For the length of hospitalization, the meta-analysis of three clinical trials (134 participants) showed a significant benefit with the use of CSF compared with the control group (WMD = - 1.58, 95% CI - 3.00 to - 0.15; p = 0.03). The combined analysis of five studies showed a significant reduction in the incidence of infections in the CSF group (45 episodes with 476 cycles) compared with the control group (91 episodes with 482 cycles) (RR = 0.56, 95% CI 0.39 to 0.80, p = 0.002). Among the adverse effects described were redness or swelling at the site of injection, fever, muscle aches, skin rash, urticaria, and bone pain. According to the evidence, the authors of this review concluded that the prophylactic administration of CSF reduces the length of the hospitalization and the rate of infection during the treatment. However, no evidence demonstrated that the use of CSF reduces the duration of neutropenia episodes or delays cycles of treatment in ALL pediatric patients receiving myelosuppressive chemotherapy.

Recently, Cheuk et al. conducted a systematic review, which objective was to compare the effect of medical interventions on the cardiotoxicity induced by anthracyclines in childhood cancer patients and survivors.¹⁵ The review included two RCTs with 203 participants, of which 135 were childhood cancer survivors and 68 were children with acute leukemia. In one of the clinical trials, enalapril was compared with placebo, whereas a 14-day treatment of phosphocreatine and vitamin C combined with adenosine triphosphate (ATP), vitamin E, and coenzyme Q10 was assessed in the other trial. In the latter trial, children showed symptomatic cardiotoxicity. For overall survival and mortality due to heart failure, this review reported that there were no deaths in neither of the compared groups nor the trials. The enalapril clinical trial reported on the occurrence of clinical heart failure, which was one case (1%) in the intervention group and six cases (9%) in the placebo group (RR 0.16, 95% CI 0.02 to 1.29; p = 0.09). Both studies reported the emergence of adverse events; dizziness, hypotension, and fatigue presented a greater incidence in the enalapril group. In the phosphocreatine study, the type of events evaluated was not specified. In the enalapril trial, no statistically significant differences were detected in the rate of change of the parameters of heart function tests between comparison groups. Moreover, in the phosphocreatine trial, all participants had normal echocardiograms before and at the end of the treatment. None of the studies reported data on the duration of the hospital stay for heart failure. The review concluded that although there is some evidence that enalapril temporarily improves some parameters of cardiac function, it is not clear whether it improves clinical outcomes. Enalapril was associated with higher risk of dizziness and hypotension. Therefore, the possible benefits with these side effects in childhood cancer survivors with asymptomatic anthracycline-induced cardiotoxicity should be evaluated. Regarding to the effect of phosphocreatine in childhood cancer patients with anthracycline-induced cardiotoxicity, data were limited and with a high risk of bias. Thus, it was not possible to establish any conclusion.

van As et al. (2012) evaluated medical interventions for the prevention of platinum-induced hearing loss in children with cancer.²¹ This review included three clinical trials with 149 participants of ages ranging between 0 and 22 years. In two RCTs, participants were diagnosed with osteosarcoma, and in the other study, with hepatoblastoma. The three trials used amifostine as a possible otoprotective intervention. None of the studies reported a follow-up. For the loss of the hearing or tinnitus in a study with 28 participants who received intra-arterial platinum chemotherapy, the review reported that the group that received amifostine (15 patients) presented asymptomatic or symptomatic ototoxicity, as well as 10 of the 13 patients of the control group. Data analysis from the study showed no significant differences between the comparison groups (RR 1.29, 95% CI 0.94 to 1.77; p = 0.11). In another trial with 39 participants where patients received intravenous platinum chemotherapy, 14/17 patients in the amifostine group and 15/19 patients in the control group presented symptomatic or asymptomatic ototoxicity. However, data analysis showed no significant differences between comparison groups (RR 1.04, 95% CI 0.76 to 1.44; p = 0.80). A study with 28 participants who received intra-arterial platinum chemotherapy reported that tumor response showed no significant differences between the treatment groups (RR 1.60, 95% CI 0.97 to 2.63; p = 0.06). There were 14 remissions among 15 patients from the amifostine group and seven remissions in the control group (12 patients). For adverse events, vomiting grade 3 or 4 and renal toxicity were reported. There was no difference in the number of events between different groups of treatment. The review concludes that there is no evidence from individual studies in children with osteosarcoma or hepatoblastoma treated with various analogs of platinum that demonstrates some benefit of the amifostine as an otoprotective intervention in comparison with no additional treatment. Since other studies for otoprotective medical interventions, and other types of malignant tumors do not exist, it is not possible to establish conclusions about their effectiveness in the prevention of ototoxicity in children treated with platinum.

Riley et al. (2015) developed a systematic review, which objective was to evaluate the effects of oral cryotherapy to prevent oral mucositis in patients receiving anticancer treatment.¹⁹ This review included studies that compared oral cryotherapy versus standard care, no treatment, or any other treatment to prevent oral mucositis produced by chemotherapy, radiotherapy, and targeted therapy. It included 14 RCTs with 1,316 participants aged between 8 and 85 years. However, only one study reported the inclusion of children. In eight studies, the participants presented a type of solid cancer, while the participants presented hematological cancers, such as multiple myelomas, Hodgkin lymphoma, or non-Hodgkin lymphoma in five studies. In most of the studies included in the review, pieces of ice to cool the oral cavity were used. The duration of the treatment varied according to the chemotherapy regime, being the longest of seven hours. The combined analysis of five studies showed the following results in patients with solid cancer who received fluorouracil (5FU) (444 participants). Oral cryotherapy reduced the risk of developing oral mucositis compared with the standard treatment or no treatment (RR 0.61, 95% CI 0.52 to 0.72; p < 0.00001); i.e., oral cryotherapy decreased the risk of developing oral mucositis in 39%. The combined analysis of five studies in patients who received treatment with melphalan before stem cell transplantation (270 participants) showed that oral cryotherapy reduced the risk of developing oral mucositis compared with the standard treatment or no treatment (RR 0.59, 95% CI 0.35 to 1.01; p = 0.05). Three trials showed that oral cryotherapy reduced the duration and severity of oral pain when compared with the control group. In addition, only one study reported that oral cryotherapy reduced the length of time of parenteral nutrition by 2.18 days compared with the control group (95% CI 0.03 to 4.33 days; p = 0.05). The most common adverse events reported in the study were of mild intensity, such as headaches, chills, numbness, taste disturbance, shooting pain from the teeth and coldness. The review concluded that cryotherapy could be useful in the reduction of the incidence of oral mucositis in patients that receive 5FU-based treatment for solid cancers.

Phillips et al. (2016) assessed the efficacy and safety of pharmacological interventions for the prevention of nausea and vomiting in children and young people (<18 years) that received chemotherapy.¹⁸ Thirty-four RCTs, which in turn included 2,023 participants, were included in the review, where standard antiemetic drugs versus placebo or any active intervention were compared. The trials examined a group of different antiemetic drugs, used different doses and comparators, and reported a variety of outcomes. Neither data on any outcomes beyond the first 24 hours of chemotherapy was reported, nor was the time of duration of the different treatments evaluated. According to the results of the review, it was reported that it was not possible to describe the interventions separated by groups of age, type of tumor or chemotherapy received. The results on the efficiency of the cannabinoids in comparison with antiemetic alternatives reported a beneficial effect of tetrahydrocannabinol compared with prochlorperazine for the complete control of acute nausea (RR 20.7, 95% CI 17.2 to 36.2) and complete control of vomiting (RR 19.0, 95% CI 13.7 to 26.3). Similarly, another trial that compared nabilone versus domperidone demonstrated a reduction of nausea. Two trials examined steroids as antiemetics; one of them assessed the efficacy of dexamethasone versus metoclopramide, demonstrating that dexamethasone was significantly better for the complete control of vomiting (RR 2.10, 95% CI 1.77 to 2.50). The other trial evaluated methylprednisolone compared with chlorpromazine for the control of vomiting, but no statistically significant differences among treatments were found (RR 1.0, 95% CI 0.54 to 1.86). Three studies comparing ondansetron with granisetron were identified; the combined analysis of the data found no statistically significant differences between the treatments for the prevention of acute nausea or vomiting retardation. However, regarding the prevention of acute vomiting, granisetron was significative better that ondansetron (RR 2.26, 95% CI 2.04 to 2.51). Four other studies compared different doses of granisetron (10 μg/kg versus 40 μg/kg); nevertheless, the results demonstrated no significant differences between doses for complete control of acute vomiting (RR 0.88, 95% CI 0.70 to 1.10). The review also reported the results of seven studies that evaluated different agents, such as chlorpromazine, metoclopramide, hydroxyzine, fentanyl, combined formulas of drugs (‘cocktails’) that contained lorazepam, dexamethasone, metoclopramide, and benztropine (LDMB), or granisetron, dexamethasone, midazolam, and diphenhydramine (GDMD), and a study that evaluated an intervention with traditional Chinese herbal medicine. From these studies, two showed that both chlorpromazine and domperidone are more effective than metoclopramide for the control of nausea and vomiting. The most common adverse effects reported with the use of 5-HT3 antagonists were sedation, somnolence, headaches, and abdominal pain, and with cannabinoids were drowsiness and alteration of mood. The review concluded that 5-HT3 antagonists (granisetron or palonosetron) were effective for the control and the prevention of nausea and vomiting in patients that receive emetogenic chemotherapy. Furthermore, adding dexamethasone improved the control of vomiting; however, adjunctive steroid risk-benefit profile remains uncertain.

A review of Ward et al. (2015), whose objective was to determine the effects of nutritional support in children and young people with cancer undergoing chemotherapy, included 14 RTC with a total of 562 participants under 21 years of age.²² The studies included children and young participants with leukemia or solid tumors that received chemotherapy, radiotherapy or both. The studies evaluated different interventions such as parenteral nutrition (peripheral or central parenteral nutrition); enteral nutrition (usual or nasogastric food intake); fluid therapy; energy-dense fructooligosaccharide (fibre) supplementation; glutamine supplementation of enteral or parenteral nutrition; value of different lipid formulations, and calory and nutritional intake, in order to establish different comparisons. For the parenteral nutrition (PN) compared with enteral nutrition (EN) (usual food intake) outcomes, a clinical trial with 25 participants reported that the PN group had an average increased weight of 12.9% in comparison with the usual food intake group. A trial with ten participants found changes in weight in favor of the NP group (mean difference 4.12, 95% CI 1.91 to 6.33; p = 0.0003). Another clinical trial with 23 participants, where measures were taken from the beginning to the end of radiation therapy, found that the mean percentage of change of the cutaneous or triceps skinfold thickness increased 13.9% in the PN group in comparison with the usual food intake group where no change was observed. The same trial reported that the mean percentage of change in the arm circumference increased 2.65% for the PN group compared with no changes in the usual food intake group. Finally, for the arm muscle circumference, the mean percentage of change decreased 0.2% in the PN group in comparison with an increase of 0.4% in the usual food intake group. The adverse events reported in these studies were sepsis, pneumonia, infection of the urinary tract without significant differences in the number of infections presented in either of the comparison groups. For mortality at the end of the study, the combined analysis of three trials with 52 participants reported no differences in the number of deaths between the PN group compared with the usual food intake group (RR 1.19, 95% CI 0.32 to 4.39; p = 0.80). For the rest of the outcomes, no significant differences were observed in the various comparisons of the trials. The review concluded that the evidence of individual studies is limited and there are no tests to suggest that PN is more efficient that EN in well-nourished children and young people with cancer undergoing chemotherapy. There is limited evidence suggesting that enteral or parenteral glutamine supplementation has no significant effect on the mucositis, infection rates or length of hospital stay. Evidence of other comparisons in patients with malnutrition remains unclear.

The above results should be interpreted cautiously since they mostly derive from low-quality evidence based on few studies and small-sized samples.

4.1. Implications for clinical practice

This review includes very recent data and even some unpublished data. The information presented offers some guidance that allows supporting the clinical practice of health care professionals in addition to strengthening the international recommendations on the use of interventions for the management of patients with cancer.

4.2. Implications for research

Further research focused on children populations is required, especially in childhood cancer. More long-term studies with representative sample sizes of this population to assess the effects of medical interventions for the management of cancer treatment-induced symptoms should be undertaken. It is essential to make trials with a greater statistical power and an adequate methodological quality.

So far, there is not enough information about the efficiency of the interventions for the prevention and management of the treatment-induced symptoms in children with cancer. The results show that only a small number of clinical trials have evaluated these interventions.

The current evidence is limited and of low quality, for which it is not possible to determine the security, efficiency, and utility of the interventions for the prevention and management of the treatment-induced symptoms in children with cancer conclusively.

In our opinion, at present, this review is the most comprehensive proof on this topic, and while there are some important implications for future research, these findings provide timely information to guide the clinician on the effectiveness of interventions, and thereby contribute to the proper use of these interventions.

Funding

This project did not receive any funding.

Conflict of interest

The authors declare no conflicts of interest of any nature.