Elsevier

European Journal of Cancer

Volume 76, May 2017, Pages 134-143
European Journal of Cancer

Review
Influenza vaccination in adult patients with solid tumours treated with chemotherapy

https://doi.org/10.1016/j.ejca.2017.02.012Get rights and content

Highlights

  • Influenza A and B infection causes significant morbidity and mortality in adult patients with solid tumours.

  • Guidelines recommend annual vaccination with the inactivated influenza vaccine (IIV) in patients treated with chemotherapy.

  • IIV reduces pneumonia, mortality and results in fewer interruptions of chemotherapy in patients with colorectal carcinoma.

  • Serological efficacy of IIV was demonstrated in 20 studies in adult patients with solid tumours receiving chemotherapy.

  • The minimum level of antibodies associated with protection (seroprotection) can be achieved in the majority of patients.

  • No safety concerns have been identified with the inactivated influenza vaccine in cancer patients.

  • IIV is an effective and safe measure to reduce influenza-related complications in adult patients regardless of chemotherapy.

Abstract

Patients with solid tumours receiving chemotherapy are at risk for influenza complications. Yearly influenza vaccination is recommended to patients treated with chemotherapy. However, adherence to vaccination is low, most likely due to lack of data on efficacy, optimal timing and safety of vaccination. There is scarce evidence for the effectiveness of the influenza vaccine in adult patients with solid tumours and chemotherapy on reduction of pneumonia, decreased mortality and fewer interruptions of oncological treatment. A review of 20 non-randomised serological studies in adult patients with different cancer types and chemotherapy provides insight in general trends of response to vaccination. Overall, the magnitude of the antibody response after influenza vaccination (i.e. seroconversion) can be lower than in healthy controls, but the majority of patients with solid tumours is able to mount a timely, protective immunological response (i.e. seroprotection) regardless of chemotherapy schedule, similar to healthy controls. Small sample sizes, patient heterogeneity and lack of comparable study designs limit more specific recommendations related to cancer type and optimal timing of vaccination. The inactivated influenza vaccine is safe to administer to immunosuppressed patients; side-effects are similar to those in healthy individuals. Although vaccination before start of chemotherapy is preferred to ensure optimal protection in adults with solid tumours, also vaccination during chemotherapy can reduce influenza-related complications considering the overall trends in serological response. Given the increased morbidity and mortality of influenza, influenza vaccination should be advocated as an inexpensive and safe preventive measure in patients with solid tumours receiving chemotherapy.

Introduction

Influenza is an acute respiratory tract infection caused by influenza virus subtypes A, B or C. Influenza A and B cause seasonal epidemics affecting 2–10% of the global population resulting in 250,000–500,000 deaths annually [1], [2]. Influenza A viruses are further classified based on the subtypes of their two surface glycoproteins: the haemagglutinin (H) and neuraminidase (N). The haemagglutinin and neuraminidase display continuous antigenic variation due to accumulation of mutations in their antigenic sites (‘antigenic drift’), which partly overcome humoural immunity in previously infected or immunised subjects. Introduction of a novel antigenically distinct influenza virus (‘antigenic shift’) renders most individuals susceptible due to lack of prior antigenic exposure. The two genetic lineages of influenza B viruses, the B/Yamagata and B/Victoria lineage, are also subject to antigenic drift. Although complication rates are low in immunocompetent individuals, influenza is a major cause of outpatient medical visits and lost productivity from missed workdays [1]. However, in immunocompromised persons, the severity of the disease and the risk of complications are substantial [3]. A fatality rate of 9–10% has been reported in cancer patients hospitalised for serious influenza-related infections [4], [5]. Various factors influence mortality: bacterial co-infection, haematologic and colorectal malignancies and lung cancer, advanced age and comorbid conditions including diseases of the heart, respiratory system, kidneys and liver and diabetes [5]. Lymphocytopenia and neutropenia in patients with solid tumours are associated with increased mortality risk [4]. Additionally, influenza virus infection and its complications result in the delay of oncological treatment [6]. Prevention of disease by seasonal influenza vaccination is recommended in cancer patients [7], [8], but adherence to this advice is low due to lack of awareness of recommendations, fear of side-effects, lack of data on optimal timing of vaccination during chemotherapy and concerns on efficacy of vaccination [9], [10], [11]. Consequently, only half of cancer patients are vaccinated [9], [10], [11]. Concerted national actions to advocate influenza vaccination are therefore instrumental to align oncologists and primary care physicians [12].

The development of influenza vaccines was initiated in the 1930s [13]. Influenza virus specific-immunity is considered predominantly antibody-mediated with a poorly defined contribution of cellular immunity. Higher levels of strain-specific antibodies are associated with increased protection [14]. Antibodies against the haemagglutinin, induced after natural infection or vaccination, are considered the most important correlates of protection against reinfection, provided that they antigenically match the epidemic strains. Most experience has been obtained with trivalent inactivated vaccines that are used in most national influenza vaccination campaigns, containing two influenza A viruses (H1N1 and H3N2) and one influenza B virus from one of the two lineages. Vaccines are annually adjusted to match the strains that are predicted to circulate. Newly developed quadrivalent vaccines containing both influenza A viruses and both influenza B lineages may provide broader protection, but these have not yet been incorporated in most national seasonal influenza vaccination campaigns.

Haemagglutinin specific antibodies can neutralise virus by preventing attachment of the virus to its receptor on airway epithelial cells. Effective neutralisation by antibodies is only possible if the antigens of the vaccine strains and the circulating epidemic strains match antigenically and if the mucosal barriers and immune system are adequately functioning.

The seasonal inactivated influenza vaccine (IIV) is advised for all categories of immunocompromised patients in most national guidelines [15], even though the level of immunosuppression may differ substantially between patients. The use of the live attenuated vaccine is contra-indicated in these patients [16].

Vaccine efficacy is determined in placebo-controlled trials, whereas vaccine effectiveness (i.e. prevention of illness in vaccinated populations) is obtained from observational cohort studies [14]. Vaccine effectiveness is established using clinically relevant outcome measures, such as proven influenza infections, number of hospital or ICU admissions or deaths. Effectiveness varies between seasons and patient categories. In the general adult population, influenza vaccination results in prevention of clinical infection in 70–90% of individuals and prevention of influenza-related hospitalisation in 90% [17]. Centers for Disease Control and Prevention (CDC) assessment of influenza vaccine effectiveness in the US using polymerase chain reaction–based diagnosis reports estimates between 10 and 60% in real life [18]. With advanced age, the antibody response to influenza vaccination declines [19]. Very few studies exist to confirm the vaccine effectiveness in immunocompromised patients [3]. Most vaccine studies report surrogate end-points such as the acquired antibody levels after vaccination that are considered to be above the threshold of protection (i.e. seroprotection) and the quantitative rise in antibody titres after vaccination (i.e. seroconversion). The standard procedure to assess antibody response to influenza A and B is by the haemagglutination inhibition (HI) assay. The thresholds and their correlation with vaccine efficacy have been established historically in healthy adults that culminated in a required seroprotection level of ≥1:40 (HI) used in influenza vaccine studies [20]. Most influenza vaccine studies apply a seroprotection rate of 70% in the vaccinated adults (18–60 years) as a minimal requirement for efficacy in the yearly changing influenza vaccines [21]. Whether these cutoff titres result in the same level of protection in immunocompromised subjects is unknown, because other essential components of the immune system may be affected by the underlying disease or immunosuppressive treatment.

In order to determine whether seasonal influenza vaccination in adult patients with solid tumours during chemotherapy is efficacious and safe, PubMed was searched using Medical Subject Headings (MeSH) terms ‘influenza vaccines’ and ‘antineoplastic agents’, without restrictions to date of study. Only papers published in English were reviewed. We only included studies on vaccination with IIV without adjuvant, since these vaccines are mostly used in national influenza vaccination campaigns. No randomised efficacy studies, but eight serological studies after administration of IIV were found for this patient category. In the references of these studies and using the option ‘similar articles’ another 12 serological studies in patients with different tumour types were found dating back to the 1970s. Twelve of the 20 studies included both cancer patients with and patients without chemotherapy that allowed comparison of the response to vaccination between cancer patients [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33] (Table 1). Five of these studies also included healthy controls. In another eight of the 20 studies response to vaccination was determined during, immediately before or after administration of chemotherapy, with four of eight also providing data from controls [34], [35], [36], [37], [38], [39], [40], [41] (Table 2). Patients were vaccinated on different moments during cycles of chemotherapy, and in 8 of 20 studies, the effect of the timing of vaccination on the vaccine response was examined. In addition, one retrospective study about the clinical effectiveness of influenza vaccination was identified in patients solely with solid tumours receiving chemotherapy [42].

Clinical effectiveness of influenza vaccination in patients exclusively with solid tumours receiving chemotherapy was evaluated in one retrospective study [42]. This retrospective cohort study in 1225 patients (1577 person years) with colorectal cancer receiving chemotherapy compared 40% of patients who were vaccinated to the remaining 60%. Vaccinated subjects experienced pneumonia significantly less frequently (7/626 versus 33/951 person years, p = 0.004). In the multivariate analysis, influenza vaccination reduced the all-cause mortality after 1 year (hazard ratio 0.88; 95%-CI: 0.77–0.99). An additional benefit was that vaccination reduced the number of oncologic treatment interruptions. No severe adverse reactions were reported. The moment of vaccination relative to the administration of chemotherapy was not provided, precluding additional recommendations on the optimal timing of vaccination. This study was included in a Cochrane meta-analysis on influenza vaccination for preventing influenza in adults with both solid tumours and haematologic diseases, which concluded that the evidence from four studies, although weak, is in favour of vaccinating adults with cancer receiving chemotherapy [43].

Twenty serological studies examined the antibody response after IIV in adult patients with solid tumours using chemotherapy (Table 1, Table 2). Cohorts with up to 14 cancer types were included and either seroprotection or seroconversion rates, or both were reported. Geometric means of the HI titres (GMT) for each of the included virus strains in the vaccine were sometimes provided. Nine studies compared responses in healthy controls, although only 3 studies used age-matched controls [30], [32], [39]. All studies were relatively small (n = 9–147), and conclusions on significance of differences in seroprotection and/or seroconversion were not provided in all studies.

Studies reported either mean age: 50–63 years or median age: 57–64 years. From the 12 studies that assessed the influence of age on seroprotection, only one (n = 119) found a correlation between lower rates and increasing age [26].

Seventeen studies determined antibody levels at baseline. Considerable proportions of individuals were protected at baseline reflecting preexisting immunity; protection rates varied between 0 and 71% for individual virus strains.

Seroconversion was recorded in 15 studies, which showed that 8–83% of patients vaccinated during chemotherapy were able to develop a fourfold increase in HI. The vast differences between studies and between virus strains are the result of preexisting immunity, immunogenicity of the vaccine and the heterogeneity of patients such as age, type of cancer treatment over four decades and timing of vaccination.

Seroprotection rates were reported in 16 studies. Overall, influenza vaccination increased seroprotection rates. After vaccination, seroprotection rates increased but differed substantially between studies and virus strains: 20–100% in all cancer patients and 32–100% in those during cycles of chemotherapy. The required seroprotection rate of >70% after vaccination was reached in nine of the 16 studies for at least one of the vaccine strains 3–4 weeks after vaccination, which means that the majority of patients showed a timely and adequate response to vaccination, regardless of treatment and relatively old age.

Eight studies compared responses between patients with and without chemotherapy; no differences were found [22], [25], [26], [27], [28], [29], [32], [33]. Although this may support the notion that the presence of a solid tumour itself does not result in a lower response to vaccination, numbers of patients without chemotherapy were small (n = 3–45).

In nine studies cancer patients were compared to healthy controls. In the 5 studies that reported lower responses in cancer patients, mostly only seroconversion rates were provided and 3 from these studies were from the 1970s [30], [31], [33], [38], [40]. Patient numbers ranged from 31 to 81, with up to 17 cancer types. In contrast, 4 of the 9 studies that did not show lower responses in patients were all recent studies providing both seroconversion and seroprotection rates. Similar limitations to sample size (n = 9–80) and heterogeneity of patient cohorts (up to 11 cancer types) were noted [29], [32], [39], [41].

Nine studies determined the correlation between number of lymphocytes and seroprotection. No such correlation was found. One study evaluated the effect of simultaneous administration of granulocyte-macrophage colony-stimulating factor (GM-CSF) in a variety of solid tumours, which did not increase antibody levels compared to others without GM-CSF [26].

Most recommendations on optimal timing of vaccination in patients receiving chemotherapy reflect expert opinion, based on a period of 1–2 weeks required for protective antibody levels to develop after vaccination and the assumed duration of immune dysfunction caused by chemotherapy. Consequently, an interval of at least 2 weeks before start or 3 months after termination of chemotherapy is advised [44], [45]. However, these intervals have not been systematically explored in patients with solid tumours receiving chemotherapy.

Instead, patients receiving chemotherapy within the previously mentioned non-recommended intervals did not demonstrate lower response rates compared to patients without chemotherapy and in four from nine controlled studies, no differences with healthy controls were found. Although these findings appear to support the notion of sufficient residual capacity of antibody production regardless of chemotherapy in solid cancers, the limited sample sizes may have reduced the chance to detect differences.

Eight studies examined the effect of the moment of vaccination during chemotherapy by comparing response rates between two distinct moments: before and during chemotherapy; exclusively during chemotherapy; before, during or after chemotherapy [23], [25], [33], [36], [37], [38], [40], [41]. Intervals between the two moments of vaccination and schedule of chemotherapy varied widely between studies or were not provided. In five studies, no differences in seroprotection were found comparing different time points of vaccination during chemotherapy [23], [25], [33], [36], [41]. Results of three other studies were conflicting, possibly relating to cancer type, type of chemotherapy and sample size. One study from 1977 determined response to vaccination in 20 patients with 11 different types of tumour and demonstrated that vaccination at the moment of lowest number of leucocytes—day 7 and beyond after start of chemotherapy—was more effective than vaccination at day 1 of chemotherapy [40]. However, the study did not find a correlation between the number of lymphocytes and vaccination response to support that assumption. Two recent studies, each in 38 breast cancer patients, could not replicate these results [37], [38]. On the contrary, early vaccination at day 4–5 after start of a cycle of chemotherapy induced stronger antibody responses (seroprotection 45–65% and 57–86%, respectively) than vaccination at day 16 (seroprotection 33–50% and 31–53%, respectively). Leucocytes were significantly lower at day 16, but absolute lymphocyte counts were not, excluding one possible explanation of this result [38]. In a smaller number of colorectal carcinoma patients (n = 18), no such difference in vaccination response comparing these time points was found, precluding any definitive conclusions about the most preferred moment of vaccination in all cancer types during chemotherapy.

IIVs are derived from viruses that have been cultured in embryonated eggs that are subsequently inactivated (i.e. killed) and purified. Recently, adjuvanted IIVs or vaccines derived from viruses that are grown on cell lines have been approved, but most influenza vaccination campaigns continue to use IIVs without adjuvant since the 1970s. Millions of doses have been administered with an excellent safety record [16]. Discomfort at the injection site may be reported, which may last 1–3 days. Systemic symptoms such as fever after vaccination are infrequently reported. Adverse events in patients with immune disorders or immunosuppression do not occur more frequently or more intensely than in persons without [16].

The presented data demonstrate that the magnitude of antibody response, i.e. seroconversion rates, can be reduced in patients receiving chemotherapy but that in the majority of patients, seroprotection can be achieved, in similar proportions as in healthy controls. Cancer type- or chemotherapy-specific recommendations cannot be distilled from the data. More observational vaccine effectiveness studies in common cancer types are needed to inform oncologists and primary care physicians on the benefits of the IIV-associated reduction of morbidity and mortality, which may lead to an increase in vaccination rates. Prospective placebo-controlled studies in these patients will not be ethically feasible due to the existing recommendations to vaccinate all risk groups. A more solid clinical confirmation of minimal protective levels in immunocompromised patients should be derived from these observational studies.

Furthermore, the available evidence on response to vaccination is based on data from small serological studies in heterogeneous patient groups without standardised outcome measures. New adequately powered studies should compare response to vaccination in cancer patients and age-matched healthy controls to evaluate the individual contribution of underlying disease, treatment and age. Patients should be stratified per regimen of chemotherapy including the use of high doses of corticosteroids and previous influenza vaccination. Paired blood samples should be obtained to determine pre- and post-vaccination leucocyte and lymphocyte counts and to enable comparison of antibody levels, i.e. seroprotection, seroconversion and GMT for all included strains in the vaccine. Studies examining the timing of vaccine administration with regard to the schedule of administration of chemotherapy should ideally confirm or refute the current expert-based recommendation of at least 2 weeks before and at least 3 months after chemotherapy. Preferentially, the response of vaccination during the period of the chemotherapy-induced drop in leucocytes should be compared to other moments before, during or after chemotherapy.

When vaccine effectiveness studies and the new serological studies consistently fail to demonstrate the efficacy and benefits of IIV in specific cancer types, other options for vaccine formulation should be explored such as repeated vaccinations, increased doses or adjuvanted vaccines [46]. These formulations are subject of investigation or have been introduced sparsely but have not yet been included in European national influenza vaccination guidelines due to paucity of data.

Future studies should also include the determination of response to vaccination in patients receiving immunotherapy that should specifically address safety concerns. Immune enhancement, such as induced by immune checkpoint inhibition and tyrosine kinase inhibition (TKI), may possibly increase the incidence of local swelling or systemic adverse events such as fever. The first study on influenza vaccination in TKI-treated patients did not report serious adverse events [32], but this requires confirmation. If it is established that adequate protective antibody levels can be achieved with influenza vaccination also during immunotherapy, patients can be informed about benefits and risks.

Section snippets

Conclusion

An elevated risk of influenza virus infection and related complications has been noted in patients with solid tumours receiving chemotherapy. Vaccination can reduce this risk, although vaccine effectiveness in this patient population has only been demonstrated to a limited extent. Current evidence of benefits showed a reduction of risk of pneumonia and decreased all-cause mortality. Fewer interruptions of cancer treatment occurred in the vaccinated persons.

Studies in the last 40 years have

Role of funding source

None.

Authors' contributions

A.V. participated in writing and data-analysis. I.S., L.S., W.O. and G.R. participated in revision and writing. H.G. participated in revision and data-analysis.

Conflict of interest statement

None declared.

References (46)

  • R. Gasparini et al.

    Clinical and socioeconomic impact of seasonal and pandemic influenza in adults and the elderly

    Hum Vaccin Immunother

    (2012)
  • Vaccines against influenza WHO position paper – November 2012

    Wkly Epidemiol Rec

    (2012)
  • R.F. Chemaly et al.

    A multicenter study of pandemic influenza A (H1N1) infection in patients with solid tumors in 3 countries: early therapy improves outcomes

    Cancer

    (2012)
  • C.D. Cooksley et al.

    Epidemiology and outcomes of serious influenza-related infections in the cancer population

    Cancer

    (2005)
  • CDC....
  • National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Prevention and treatment of...
  • M.L. Wumkes et al.

    Influenza vaccination coverage in patients treated with chemotherapy: current clinical practice

    Neth J Med

    (2013)
  • C. Hannoun

    The evolving history of influenza viruses and influenza vaccines

    Expert Rev Vaccines

    (2013)
  • J.J. Treanor

    Clinical practice. Influenza vaccination

    N Engl J Med

    (2016)
  • ...
  • L.A. Grohskopf et al.

    Prevention and control of seasonal influenza with vaccines

    MMWR Recomm Rep

    (2016)
  • A.E. Fiore et al.

    Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP)

    MMWR Recomm Rep 2008

    (2008)
  • CDC. CDC....
  • Cited by (29)

    • To Vaccinate or not: Influenza Virus and Lung Cancer Progression

      2021, Trends in Cancer
      Citation Excerpt :

      Additionally, the appearance of malignant cancer is linked to the prevalence of influenza infection that occurred 1 year before the detection of cancer [2], and a higher risk of developing lung cancer is related to cumulative exposure to the virus [3]. Recent studies have suggested that treatment with antivirals or vaccination against influenza may provide clinical benefit, including reduced hospitalization and reduced mortality, to patients with cancer [4]. Current recommendations from the Centers for Disease Control and Prevention (CDC) and American Society for Clinical Oncology (ASCO) are to vaccinate patients with cancer (https://www.cdc.gov/cancer/flu/basic-info.htm; https://ascopost.com/issues/september-25-2019/flu-vaccine-effectiveness-in-patients-with-cancer).

    • Lung cancer

      2020, Revue des Maladies Respiratoires Actualites
    View all citing articles on Scopus
    View full text