Protecting Pediatric Oncology Patients From Influenza

This review examines evidence to support various strategies to protect pediatric oncology patients from influenza-related morbidity. Influenza vaccination should be considered standard. Additional evidence-supported measures include antiviral treatment, antiviral prophylaxis, cohorting of patients, and hospital infection control measures.

Learning Objectives

1. Identify optimal vaccination strategies and define the vaccine response rates among pediatric chemotherapy patients.

2. Explain the advantage of beginning empiric antiviral therapy.

3. Describe the need for family member vaccination, hygiene measures, and social distancing.

Abstract

Influenza is a common respiratory pathogen. Its severity can be unpredictable, but people with chronic illness are at increased risk of severe infection, complications, and death from influenza. This review examines evidence to support various strategies to protect pediatric oncology patients from influenza-related morbidity. Influenza vaccination should be considered standard. Additional evidence-supported measures include antiviral treatment, antiviral prophylaxis, cohorting of patients, and hospital infection control measures. Data from other high-risk populations support the vaccination of family members, double-dose or high-dose vaccination, and the use of barrier methods. These measures have the potential to optimize patient outcomes because there will be fewer treatment interruptions for acute illness. These strategies can also protect patients from prolonged hospitalizations and morbidity related to influenza.

Implications for Practice:

Pediatric oncology patients can have very severe infections with the influenza virus. Their vulnerability is due to chronic illness and immune suppression. Simple strategies, such as ensuring vaccination, infection control measures, and education regarding personal protection can protect this population that is susceptible to severe influenza.

Introduction

Influenza is an acute respiratory infection affecting 10%–20% of the population each year in all age groups [1]. Most people with influenza exhibit an uncomplicated febrile respiratory illness and recover within a week; however, several high-risk groups have been identified in which severe complications such as pneumonia, encephalitis, and cytokine storm occur at higher rates [2]. The high-risk groups broadly include infants, the elderly, and people with chronic illness. Influenza-related deaths range from <10 per 100,000 for healthy people up to >600 per 100,000 for chronically ill adults [3].

All patients with hematologic or solid cancers undergoing chemotherapy are considered to be at high risk of influenza-related complications; children with malignancies are particularly vulnerable. A nationwide sentinel disease surveillance system in Taiwan estimated that between the years 1999 and 2005, the incidence of influenza-related hospitalizations was 82.4 per 100,000 patients <20 years old with malignancies [4]. Active surveillance for influenza-like illness conducted on pediatric and adult oncology units during the 2009 influenza pandemic revealed that 20% of patients hospitalized with fever were infected with influenza [5]. Other studies have suggested that as many as two-thirds of pediatric oncology patients who were diagnosed with influenza during that 2009 pandemic required hospitalization [6]; complications occurred in 10%–20% of those hospitalized [7–10]. This review includes recent information on the risks associated with specific pediatric oncology patient populations, preventive strategies, and management options.

Methods

A systematic review of the literature was performed. Additional studies that were known to the authors were included in some cases as substantive foundation material. PubMed was searched for the timeframe of 1990 to August 2012. The search terms included chemotherapy, influenza, vaccine, oseltamivir, oncology, neoplasm, child cancer, immunosuppression, vaccination barriers, and parental compliance with vaccination. Review articles were included where synthetic information was pertinent. In addition, the Cochrane database was examined to identify additional studies. For each article included in this review, the complete publication was examined.

Epidemiology of Influenza in Oncology Patients

Transmission

Prevention is the cornerstone of all efforts to control influenza and understanding the source of infection is critical. Influenza is easily transmitted within the health care setting. Investigations of outbreaks that occurred in pediatric oncology units have identified infected visitors [11] and infected health care workers [12, 13] as common sources of infection. These observations highlight the importance of universal influenza vaccination of health care workers as well as enforcement of appropriate use of sick leave. Multiple reports have also described the role that group activities have played in propagation of influenza outbreaks on pediatric oncology units [11, 12]. Providing protection for patients is complex because infected individuals may be difficult to identify and transmission can occur from asymptomatic individuals. The usual incubation period for influenza ranges from 1 to 4 days. The virus is typically shed in respiratory secretions 1 day before the onset of symptoms and young children can continue to shed virus for 10 or more days [14]. The duration of viral shedding in immunocompromised patients can be even more prolonged, allowing for sustained transmission in hospitals [7, 15].

Risk Factors for Severe Influenza

Lymphopenia, not neutropenia, was commonly detected in children with cancer who were hospitalized with influenza [8]. An analysis of a series of 27 patients with laboratory-confirmed influenza and malignancy demonstrated that chronic steroid use and delayed initiation of oseltamivir were associated with influenza-related lower, as compared with upper, respiratory tract infection [16]. Additional risk factors include recent administration of chemotherapy and a low absolute lymphocyte count (usually <500 cells/μL) [7, 8, 17, 18]. Studies performed in adult oncology patients have demonstrated that influenza A and hematologic malignancies are both independent risk factors for severe infection [19].

Several case series of children with malignancy who were infected with 2009 pandemic influenza reported that up to 20% of children with malignancy developed severe disease when infected with this strain [19–21]. Elbahlawan et al. reported that 5 of 28 patients hospitalized with pandemic influenza developed acute respiratory failure [22]. Due to the unique epidemiology of this strain of influenza, it is difficult to generalize these findings to years in which other strains of influenza are in circulation. Although vaccination is effective, severe disease has occurred in vaccinated patients, suggesting that the most immunologically vulnerable patients may not respond optimally to vaccines [17].

Clinical Features

Hospitalization

Respiratory viral infections in children undergoing cancer therapy are seen more frequently and are more severe than in healthy children [18]. Studies done prior to implementation of routine yearly influenza vaccinations showed that children with cancer are more likely to contract influenza than healthy children in similar environments [23].

Children receiving cancer therapy are more likely to be hospitalized than healthy children when they contract influenza. Up to two thirds of such children are admitted to the hospital, prompted primarily by occurrence of fever [7–9]. These hospitalizations last from 2 to 7 days on average [8]. Most patients are febrile at presentation, and fevers can last 1–2 weeks [8].

In addition to frequent hospitalization, influenza in children with cancer causes significant clinical morbidity. Severe respiratory complications, including respiratory failure, pneumonia, and need for ventilatory support, occur in 10%–20% of these hospitalized children [8, 10, 17]. Less severe respiratory complications, such as hypoxia and need for bronchodilators, are common [8, 17]. A small number of children require prolonged oxygen therapy after the acute illness has subsided [8].

Complications

Intensive care admission occurs in up to 10% of influenza-infected children with cancer, and death occurs in up to 5% [17]. Bacteremia as a complication of influenza infection in these children occurs in about one in six children. Pathogens as Pseudomonas spp., Enterobacter spp., Streptococcus spp. and coagulase-negative staphylococci [6, 8] have been described. Secondary bacterial infections may also occur as clinical infections, such as pneumonia or otitis media [10]. Children infected with influenza during the 2009 pandemic were reported to be more likely to develop pneumonia when they were neutropenic [7]. Less common complications of influenza infection include fungal pneumonias (e.g., Aspergillus pneumonia), other invasive fungal infections, and hemophagocytic lymphohistiocytosis [24–26].

For children with cancer, influenza infection may last up to twice as long as in healthy children [10, 23]. In addition, viral shedding can occur for up to 6 weeks. Canadian children who were infected with the 2009 pandemic influenza continued to shed virus for a median of 46 days [7].

Impact on Chemotherapy

Illness due to influenza in children with cancer causes significant delay in chemotherapy or other cancer-specific therapy in 20%–80% of patients [7, 8, 10, 23]. In one population of children with 2009 pandemic influenza infection, 55% of the children experienced treatment delays averaging 3 weeks; the longest delay in this group was 43 days. Others have also reported average treatment interruptions of 3 weeks or more [6, 8]. The impact of treatment interruptions due to influenza specifically is unknown; however, emerging evidence suggests that relapse rates are higher among children with acute lymphoblastic leukemia who do not receive full maintenance chemotherapy [27]. Studies of other malignancies, primarily in adults, suggest that interruptions are deleterious but the specific outcomes related to treatment delays have not been defined for most types of malignancy [28–32].

In one population of children with 2009 pandemic influenza infection, 55% of the children experienced treatment delays averaging 3 weeks; the longest delay in this group was 43 days. Others have also reported average treatment interruptions of 3 weeks or more. The impact of treatment interruptions due to influenza specifically is unknown; however, emerging evidence suggests that relapse rates are higher among children with acute lymphoblastic leukemia who do not receive full maintenance chemotherapy.

Treatment of Influenza

There are two antiviral medications that are approved by the U.S. Food and Drug Administration and recommended for treatment or chemoprophylaxis of influenza [33–36]. Oseltamivir and zanamivir have activity against influenza A, influenza B, and 2009 pandemic influenza virus. Oseltamivir has been approved in patients older than 2 weeks. Zanamivir has been approved for chemoprophylaxis in children older than 5 years and for treatment in children older than 7 years. Treatment with antiviral medications has been shown to decrease fever duration, reduce the risk of complications from influenza, and shorten the length of hospital stay in the general population. Because of resistance, treatment with amantadine or rimantadine is not recommended.

Early treatment (ideally within 48 hours) is recommended for patients with confirmed or suspected influenza who are hospitalized, have severe or progressive illness, or are at high risk for complications. Treatment should not wait for laboratory confirmation of disease in patients who are at high risk of complications. Patients can have multiple risk factors, which may further increase the risk of morbidity: age younger than 2 years, residence in a chronic care facility, immunosuppressive medications, human immunodeficiency virus, and chronic disease including pulmonary, cardiovascular, renal, hepatic, hematological, neurologic, and metabolic conditions; diabetes; mental retardation or severe developmental delay; muscular dystrophy; and spinal cord injury.

Few studies have evaluated the impact of antiviral therapy specifically for the treatment of influenza in patients undergoing cancer treatment or after bone marrow transplantation (BMT). One study showed BMT patients with 2009 pandemic influenza had prolonged viral shedding (median: 46 days) after symptom resolution; longer treatment was correlated with shortened duration of viral shedding [7]. In another cohort of 62 patients over 12 consecutive influenza seasons, early antiviral therapy was suggested to contribute to decreased progression to pneumonia and decreased viral shedding [37]. Patients were more likely to develop pneumonia if infected earlier after transplantation. In another cohort of BMT patients receiving oseltamivir within 2 days of symptoms, only 5% developed pneumonia and none died from influenza [38].

For non-BMT patients, advantages of early antiviral treatment are similar to those in healthy children [39]. Patients treated with oseltamivir had significant reductions in the risk for respiratory illnesses (other than pneumonia), otitis media, and hospitalization. Delays in treatment were associated with increased progression to lower respiratory tract infection and mortality in this population [40, 41].

Almost all strains of influenza are susceptible to oseltamivir and zanamivir [42]; however, there are a few case reports of resistant influenza isolated from immunocompromised patients. Ison et al. reported three cases of immunocompromised patients with influenza with molecular markers of resistance to anti-influenza drugs [43]. Another study found that half of patients receiving chemotherapy had oseltamivir-resistant influenza viruses with resistant strains identified both prior and during therapy [44].

Influenza Vaccination for Children on Chemotherapy

Although the Advisory Committee on Immunization Practices and the American Academy of Pediatrics have consistently recommended influenza vaccination for children undergoing treatment for malignancy, there has traditionally been little acceptance of this recommendation by oncologists [45–47]. Although overall humoral immunity is compromised, vaccine responses still occur and vaccination is a critical strategy [48–50]. Many hospitals now include influenza vaccination rates among high-risk populations as a quality measure, which would be expected to improve the overall vaccination rates among children on chemotherapy.

Beginning with two landmark studies in the 1970s, all the studies to date have defined serologic responses to the influenza vaccine as the outcome metric [51–58]. Although serologic responses are a critical measure, there is no consensus on what level of antibody should be considered protective in an immunocompromised population [59–61]. Hemagglutination inhibition titers of 1:40 confer protection in the majority of healthy young adults [62–64]. However, in a study of vaccine efficacy in pediatric oncology patients, there was evidence of breakthrough infection in individuals with pre-existing titers of 1:320 [65]. In spite of universal vaccination in the study population, a breakthrough infection rate of 15% using serologic conversion was found. Using positive polymerase chain reaction testing as a second ascertainment strategy, a breakthrough infection rate of approximately 8% was found [65]. These studies suggest that a high rate of protection is conferred by vaccination but also reveal that there is still a need to further optimize management.

A total of 10 studies were identified that included primary data on influenza vaccine efficacy in children undergoing chemotherapy since 1990 [51, 52, 54, 56, 57, 66–70]. These studies generally included patient populations that were heterogeneous in terms of types of malignancy, patient age, and phase or time on chemotherapy; however, they all generally reached the same conclusions. Nearly all the studies found that the patients were generally capable of mounting a response, although the magnitude of the response was generally lower than in healthy controls. A summary of the recent studies is given in Table 1. Two reviews of primary data led to the common recommendation to vaccinate while on maintenance chemotherapy. In these studies, patients were found to have protective anti-influenza antibody rates of 30%–70% (titers of at least 1:32 or 1:40) [52, 70]. Seroconversion rates of 30%–80% (a fourfold increase) were seen in patients [51, 54, 56–58, 66–70].

Table 1.

Recent pediatric oncology vaccine studies

Abbreviations: ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; BMT, bone marrow transplantation; IgG, immunoglobulin G; LAIV, live attenuated influenza vaccine; TIV, trivalent inactivated influenza vaccine.

It has been presumed that the type and phase of chemotherapy could impact vaccine response [69]. One recent study directly compared responses across tumor types. This study found that subjects with acute myelogenous leukemia had extremely limited serologic responses to the vaccine (Table 2). Surprisingly, patients with acute lymphocytic leukemia had the best serologic response early in the treatment protocol [65]. In children with solid tumors, responses were limited but did not appear to vary with the aggregate time on chemotherapy. A reasonable strategy is to vaccinate children with cancer as soon as the vaccine is available. Evidence also supports giving multiple doses of the vaccine. Although trials are limited in children, this has been shown in other settings to boost vaccine serologic responses [71, 72].

Table 2.

Seroconversion among patient subgroups

Data are from [65, 104, 105].

Vaccine Refusal

Parents of healthy children who are not interested in vaccination for their children consistently describe several reasons. Many parents are concerned about side effects, feel that their child has a low risk of disease, or feel that the vaccine causes disease [73]. Other parents are not interested after soliciting advice from their social network [74]. For patients with chronic illness, one of the strongest predictors for vaccination is physician recommendation [75–78]. Other reasons for vaccinating chronically ill children include relative recommendation, easy access to physician office or reminder, belief in vaccine efficacy, and higher parental educational level [79]. Not surprisingly, parents are more interested if they feel that the underlying illness is severe or the vaccine will lessen symptoms or exacerbations from the chronic disease [77, 79].

Kersun et al. published their experience in surveying 100 consecutive parents in an outpatient oncology clinic [80]. The children had a range of oncologic diagnoses and almost all were currently receiving chemotherapy. In all, 94% were current with their immunizations prior to diagnosis and 85% planned to receive the influenza vaccine during the upcoming season. However, 30% were worried their child would get influenza from vaccine, 23% worried their child was too ill, and 56% were concerned about side effects. Of those surveyed, 11 parents stated their child would not receive vaccine, although almost all were current on immunizations prior to diagnosis. More than half of these parents were not confident in the vaccine and felt it was unnecessary. These data highlight the importance of education in the comprehensive approach to influenza control.

Alternative Prevention Strategies

Additional strategies can help to reduce the risk of influenza infection in children with malignancies, including chemoprophylaxis and other behavioral interventions. These measures should be considered adjuncts to vaccination and should not be used in lieu of vaccination.

Primary chemoprophylaxis can be an effective alternative to vaccination in patients who cannot receive influenza vaccination or who cannot respond to influenza vaccine (e.g., a severe primary or secondary immunosuppressing condition). A recent Cochrane review demonstrated that chemoprophylaxis was approximately 90% effective in preventing influenza A in otherwise healthy children. However, it is estimated that 17 children would need to receive chemoprophylaxis to prevent a single case of influenza.

Primary Chemoprophylaxis

Primary chemoprophylaxis refers to the use of antiviral medications to prevent influenza infection. Various chemoprophylaxis strategies have been shown to be effective and are recommended for high-risk patient populations who cannot be vaccinated, cannot respond to vaccination, or have not yet received the vaccine [81]. Seasonal chemoprophylaxis refers to the daily administration of an anti-influenza medication for the 14- to 18-week period when local influenza virus circulation is at its peak. Because of relatively high rates of resistance in circulating strains of influenza A, amantadine and rimantadine are not currently recommended for chemoprophylaxis against influenza [42]. Primary chemoprophylaxis can be an effective alternative to vaccination in patients who cannot receive influenza vaccination or who cannot respond to influenza vaccine (e.g., a severe primary or secondary immunosuppressing condition). A recent Cochrane review demonstrated that chemoprophylaxis was approximately 90% effective in preventing influenza A in otherwise healthy children [82]. However, it is estimated that 17 children would need to receive chemoprophylaxis to prevent a single case of influenza.

There are two reports of chemoprophylaxis in pediatric oncology patients. One study described an influenza outbreak (n = 6) in a pediatric oncology unit with successful chemoprophylaxis of all exposed hospitalized patients [11]. Oseltamivir treatment was instituted in addition to isolation of infected patients, use of protective equipment, and visitor restriction. No new cases developed after these measures were instituted. Another outbreak of four cases of influenza A occurred in an outpatient residential facility for BMT patients [83]. Oseltamivir prophylaxis was given to 45 patients. No new cases of influenza A developed after initiation of oseltamivir. There is one study evaluating oseltamivir prophylaxis during influenza season for pediatric patients with cancer and BMT [84]. Patients receiving chemotherapy or BMT (n = 32) were given oseltamivir prophylaxis for 8 weeks. None of the patients developed influenza.

Secondary Chemoprophylaxis

Secondary chemoprophylaxis refers to the use of antiviral medications for a relatively short period after a known or suspected exposure. Studies performed in households in which a person with a documented influenza infection resides have revealed that antiviral medications can prevent secondary cases of influenza [85, 86]. These studies have not been performed in households with immunocompromised children. Oseltamivir prophylaxis has been recommended in the setting of a nosocomial outbreak of influenza on an oncology unit to interrupt further transmission (as described above) [87]. When postexposure prophylaxis is used in the household setting, it is typically administered for 2 weeks. When oseltamivir is used to control an outbreak of influenza in a health care setting, the drug is typically continued for 1 week after the last documented case of nosocomial influenza.

Barrier Protections

Appropriate application of standard and transmission-based precautions is important to prevent patient-to-patient transmission of influenza. In the hospital setting, transmission-based precautions are recommended for any patient with known or suspected influenza. For most institutions, droplet precautions are used, although some institutions use droplet and contact precautions for respiratory viral infections due to the multiple modes through which these viruses can be transmitted. Cohorting or segregating infected patients to limit their contact with uninfected patients is also an important component of influenza control and can be applied in either an inpatient or outpatient setting [88]. Although sometimes used by families during outings to group activities (e.g., parties, shopping), there have not been any systematic studies of mask use in public to protect oncology patients from infection with influenza.

Social Distancing and Other Interventions

The Centers for Disease Control and Prevention [89, 90] recommends that oncology patients follow “good health habits” as one component of preventing seasonal influenza. These practices include such measures as avoiding close contact with people who are sick, frequent hand cleaning, and avoiding self-inoculation by not touching the mucus membranes of the eyes, nose, and mouth. Additional precautions include the early identification and separation of patients with symptoms of influenza-like illness from other patients.

Conclusions

Pediatric oncology patients are at increased risk of severe influenza. A unified, global strategy can minimize the impact of influenza on treatment and limit patient morbidity and mortality. Vaccination is underused and is probably the single most important intervention [45]. Vaccination is generally effective and should be the cornerstone of all prevention strategies. Vaccination of family members is an adjunct strategy that could be more widely used [91].

To protect oncology outpatients during influenza season, the following outpatient practices are recommended:

1. Prescreen all scheduled patients for influenza-like illness symptoms.

2. Instruct patients with influenza-like illness symptoms to don a mask upon arrival to the clinic.

3. If possible, place patients with influenza-like illness into a private examination room upon arrival.

4. If immediate placement in a private examination room is not possible, segregate patients with influenza-like illness symptoms within the waiting room to avoid mingling of infected and uninfected patients [89].

In the inpatient setting, droplet precautions and chemoprophylaxis for at-risk patients is warranted. The other component of optimal treatment is the approach to infection control, including specific hospital guidelines, social distancing, and hygiene measures in the home and a concerted effort at the early recognition of influenza in the patient and in others in the hospital setting. Table 3 outlines evidence-supported measures to reduce influenza in pediatric oncology populations.

Table 3.

Evidence-based interventions to minimize influenza morbidity and mortality

Postexposure prophylaxis has demonstrated efficacy in preventing disease in exposed household or other close contacts. Postexposure prophylaxis is a reasonable practice for high-risk populations if the exposure is known within 48 hours. Use of antiviral medications is critical when there is an institutional outbreak of influenza among patients at high risk for complications secondary to influenza. For active infections, early treatment should be instituted.

It is important to understand similarities and differences with the adult oncology population. Adult oncology patients also benefit from vaccination, although insurance companies do not uniformly cover costs [92–96]. Several studies have implicated rituximab-based protocols as associated with impaired vaccine responses; otherwise, specific protocols have not seemed to affect vaccine responses [97, 98]. Patients with solid tumors, as was seen in children, seem to have slightly better responses than those with hematologic malignancies [99–101]. Early vaccination seems to be optimal for adults as was seen in children [102]. Also concordant with the pediatric experience is the high rate of morbidity and mortality in adult cancer patients with influenza. A case fatality rate of 23% was seen in BMT recipients [104]. In non-BMT patients, case fatality rates have ranged from 10%–30% [2]. Therefore, recommendations for adult patients are largely concordant with those for children.

This review summarizes the current knowledge regarding optimal protection strategies for children with cancer. As survival rates from cancer continue to improve, attention to comorbidities has increased. Optimizing infection control can have significant benefits for patients, so a comprehensive approach to influenza is recommended.

This article is available for continuing medical education credit at CME.TheOncologist.com.

Author Contributions

Conception/Design: Kathleen Sullivan, Leslie Kersun, Anne Reilly, Susan Coffin

Collection and/or assembly of data: Kathleen Sullivan, Leslie Kersun, Anne Reilly, Susan Coffin

Data analysis and interpretation: Kathleen Sullivan, Leslie Kersun, Anne Reilly, Susan Coffin

Manuscript writing: Kathleen Sullivan, Leslie Kersun, Anne Reilly, Susan Coffin

Final approval of manuscript: Kathleen Sullivan, Leslie Kersun, Anne Reilly, Susan Coffin

Disclosures

The authors reported no financial relationships.

Section Editors: Rochelle Bagatell: None; John Cunningham: None

Reviewer “A”: Sanofi, Allergan, Peregrine, Pharmacyclics (C/A); Eli Lilly (H); Gilead, Merck, Curis, Onyx, Bristol-Myers Squibb, Amgen, Celgene, Exelixis (O); Epizyme (SAB); PharmaMar (Board of Directors)

C/A: Consulting/advisory relationship; RF: Research funding; E: Employment; H: Honoraria received; OI: Ownership interests; IP: Intellectual property rights/inventor/patent holder; SAB: scientific advisory board