To date, treatment of HMPV infection is mainly supportive. Several treatment regimes have been investigated. Most of these therapeutic options, like innovative approaches based on fusion inhibitors and on RNA interference, seemed effective in vitro and in animal studies.

Ribavirin is a nucleoside with broad spectrum inhibitory activity against a variety of RNA and DNA viruses, including HMPV. Ribavirin has demonstrated in vitro inhibition of tumor necrosis factor-alfa, interferon-gamma and interleukin (IL)-10, suggesting a down regulation in Th1 and Th2 cytokine production and an increase of IL-2 production by peripheral blood mononuclear cells [43,62]. Ribavirin may terminate T-cell immune-mediated damage caused by viral infections. It limits viral transcription and showed to have immunomodulatory effects [63]. The in vitro results are confirmed by an in vivo study in BALB/c mouse [64,65,66,67,68,69].

Immunoglobulins for therapeutical goals can be divided in specific and non specific.

Palivizumab (Synagis®) contains humanized monoclonal antibodies that can recognize a highly conserved neutralizing epitope on the fusion protein of RSV. It showed to have preventive effects in infants at high risk of severe hRSV infections; monthly palivizumab injections reduced RSV hospitalizations by 50% compared with placebo [70,71,72,73]. Motavizumab is another RSV specific monoclonal antibody preparation, which is developed after the success of palivizumab. It showed to be non-inferior to palivizumab for prevention of RSV hospitalization in high-risk children [74]. These data of effectiveness of humanized monoclonal Abs against RSV infection has prompted a similar approach for protection against HMPV.

MAb 338 is one of the antibodies that was developed to target the HMPV fusion protein. It appeared effective in animal models in which it neutralized the prototypic strains of the four subgroups of HMPV, significantly reduced the pulmonary viral titer, limited severe acute manifestations and limited bronchial hyper-reactivity. In mice, it appears to have both prophylactic and therapeutic benefits [75]. Hamelin et al. also showed that it could be useful after infection and not only as preventive measure [76].

Williams et al. tested a fully human monoclonal antibody fragment (Human Fab DS7) with biological activity against HMPV in vivo and in vitro and demonstrated a prophylactic and therapeutic potential against severe HMPV infection. When Fab DS7 was given intranasally to cotton rats, a >1,500-fold reduction in viral titer in the lungs and a modest 4-fold reduction in the nasal tissues was found. A dose-response relationship between the dose of DS7 and virus titer was seen [77,78].

Wyde et al. showed that standard immune globulin preparations (thus without selection for antibodies to a particular microorganism or its toxin), initially used as preventive measure against hRSV, also inhibit replication of HMPV in vitro [64].

The combination of oral and aerosolized ribavirin with polyclonal intravenous immune globulin (IVIG) seems an effective treatment for severe HMPV infections, but no randomized controlled trials in humans have been performed. Despite this lack of good trials in human, a lot of experience has been gained meanwhile in individual cases and small case series [69,79,80].

Both ribavirin and IVIG are expensive and have disadvantages. Ribavirin is potential teratogen and administration by nebulization must be carried out via a small particle aerosol generator [81]. Therefore, in daily practice ribavirin nebulization is seldom used for HMPV infection. In addition, health care providers who are pregnant or attempting to become pregnant should avoid contact with patients receiving treatment with aerosolized ribavirin. Furthermore, IVIG requires large fluid volumes infusions, generates a high protein load and is associated with adverse side effects in children with congenital heart disease [82].

Fusion inhibitors target the first steps of the viral replication cycle. Deffrasnes and colleagues tested nine inhibitory peptides with sequences homology with the HRA en HRB domains of the HMPV fusion protein and demonstrated potent viral inhibitory activity in vitro of five of these peptides. One peptide, HRA2, displayed very potent activity against all four HMPV subgroups. BALB/c mice that received the HRA2 peptide and a lethal HMPV intranasal challenge simultaneously were completely protected from clinical symptoms and mortality [83]. The study of Miller and colleagues demonstrated that individual HR-1 peptides could lead to effective viral inhibition [84]. These peptides could be used in the prevention of severe infection in vulnerable patients after exposure, but the clinical role post-infection has to be investigated.

RNA interference (RNAi) is a recently discovered interesting approach for treatment of RNA virus infections. RNAi is a naturally occurring intracellular inhibitory process that regulates gene expression through the silencing of specific mRNAs. The small RNAs, the microRNA (miRNA) and small interfering RNA (siRNA), can down-regulate protein production by inhibiting targeted mRNA in a sequence-specific manner. RNAi therapeutics have been shown to be active in vitro and in vivo against respiratory syncytial virus, parainfluenza and influenza [85,86,87]. Deffrasnes et al. successfully identified two highly efficient siRNAs against HMPV in vitro, targeting essential components of the HMPV replication complex [88]. Very recently, Preston and colleagues designed and validate a siRNA molecule that is effective against the G gene of hMPV in vitro. Although, a significant reduction in G mRNA did not reduce viral growth in vitro or induce a significant type I interferon (IFN) response, hMPV G might still be a valid target for RNAi as G is required for viral replication in vivo [89].

Wyde et al. have also demonstrated that both the sulfated sialyl lipid (NMSO3) and heparin have antiviral activity against HMPV in vitro. NMSO3 acts most likely by inhibiting attachment and penetration of the virus and may inhibit cell-to-cell spread [90].

Several in vitro and animal studies have been performed investigating the development of an HMPV vaccine. However, no human studies have been performed yet and no vaccine is available up till now.

Results of studies performed in rodent and non-human primate models look promising, but very little research is performed in human volunteers. A variety of live attenuated, virus vectored, inactivated virus and subunit vaccines have been tested in animal models and showed to have immunogenicity and protective efficacy [91].

HMPV expresses the major surface glycoproteins F and G. Two main genetic virus lineages exist worldwide which have a similar highly conserved F (fusion) protein. Immunization strategies have been targeted against these surface proteins. Immunization with monoclonal antibodies against the F protein shows a prophylactic effect [77,78,92,93]. Several animal studies investigating the immunization with a chimeric virus vector using a bovine parainfluenza virus 3 expressing the HMPV F protein, adjuvanted soluble F protein or F protein DNA show protective immunity after a HMPV challenge [92,93,94,95,96].

Immunization with the HMPV attachment (G) glycoproteins did not show any production of antibodies or protection [97]. Ryder et al. also demonstrated that HMPV G is not a protective antigen. They evaluated the protective efficacy of immunization with a recombinant form of G ectodomain (GDeltaTM) in cotton rats. Although immunized animals developed high levels of serum antibodies to both recombinant and native G protein, they did not develop neutralizing antibodies and were not protected against virus challenge [98].

Studies investigating immunization with inactivated HMPV show an enhanced immune response with even lethal outcome following HMPV infection in animals [99,100,101]. Use of live-attenuated viruses generated by reverse genetics or recombinant proteins, tested in animals, showed encouraging results. Live vaccines are mimicking natural infection; however natural infection does only lead to transient protective immunity [38]. This makes an extra challenge for vaccine development.

The primary strategy is to develop a live-attenuated virus for intranasal immunization. Reverse genetics provides a means of developing highly characterized ‘designer’ attenuated vaccine candidates. To date, several promising vaccine candidates have been developed, each using a different mode of attenuation. The first candidate involves deletion of the G glycoprotein, providing attenuation that is probably based on reduced efficiency of attachment. The second candidate involves deletion of the M2-2 protein, which participates in regulating RNA synthesis and whose deletion has the advantageous property of up-regulating transcription and increasing antigen synthesis. A third candidate involves replacing the P protein gene of HMPV with its counterpart from the related avian metapneumovirus, thereby introducing attenuation owing to its chimeric nature and host range restriction. Another live vaccine strategy involves using an attenuated parainfluenza virus as a vector to express HMPV protective antigens, providing a bivalent pediatric vaccine [102].

As outbreaks with HMPV are frequently described, control measures to prevent HMPV transmission in hospitals and long-term care facilities seem justifiable [43,103,104,105]. When patients with HMPV infection are hospitalized, infection control measures similar to those taken in case of RSV infection should be taken including droplet isolation until clinical recovery. The Dutch working party on infection prevention advises to apply droplet isolation to all patients hospitalized with bronchiolitis until clinical recovery [106]. No specific advice is formulated for HMPV infections. The CDC advises contact and droplet precautions for infants and young children with respiratory infections; however no advice for adults is given [107]. In our hospitals (Diakonessenhuis Utrecht and the University Medical Centre Utrecht, Utrecht, The Netherlands), droplet isolation is applied to all patients with HMPV infection until clinical recovery. We do not routinely perform control RT-PCR on nasopharyngeal swabs after clinical recovery.

Respiratory viruses are a common trigger for exacerbations of COPD, and have been associated with respiratory failure in patients with cardiopulmonary disease such as COPD and congestive heart failure [108,109]. Walsh et al. performed a cohort-study during four winters to investigate the clinical outcome and incidence of HMPV infections [110]. Serum samples were taken before and after the observation period (November 15 to April 15) each year. In case of respiratory symptoms a nasopharyngeal swab for HMPV RNA analysis and serum were sampled. They showed that 71% of infections with HMPV were asymptomatic in the healthy young adults (19–40 years) in contrast to 39% in the high risk adults (patients with symptomatic lung disease, COPD, congestive heart failure). These patients were also more likely to use medical care service. Patients were ill for a mean of 10 days in the young adults versus 16 days in the high risk group. Johnstone et al. investigated the potential role of respiratory viruses in the natural history of community-acquired pneumonia (CAP). In 39% of the 193 patients who were admitted because of CAP, a pathogen was identified. Of these pathogens, 39% were viruses and the easily transmissible viruses such as influenza, HMPV, and RSV were the most common (respectively 24, 24 and 17%). There were few clinically meaningful differences in presentation and no differences in outcomes according to the presence or absence of viral infection. The patients with viral infection were, compared with bacterial infection, significant older, more likely to have cardiac disease and more frail [111].

This is in accordance with the results of Hamelin et al. who found HMPV in 4.1% of patients with CAP or exacerbation of chronic obstructive pulmonary disease [112]. Martinello et al. also showed that HMPV was frequently identified in patients hospitalized because of an exacerbation of COPD [113]. HMPV (both genotype A and B) was identified in nasopharyngeal specimens (by RT‑PCR) in 12% of these patients (6/50). RSV, influenza A and parainfluenza type 3 were identified in respectively 8%, 4% and 2%.

Along with these results, Williams and colleagues showed that HMPV was detected (by RT-PCR of nasal wash specimens) in almost 7% (7/101) of the adults hospitalized for an acute asthma exacerbation, compared to 1.3% in follow-up patients (p = 0.03). While none of these patients tested positive for HMPV three months after discharge, a direct etiologic role of the virus seems very likely [114,115].

Recently, we reported a case series of adult patients, including two patients known with COPD, with severe HMPV infections with respiratory insufficiency and the need of ICU admission [116].

Since adults are not routinely screened for HMPV in the hospital and clinical course can be asymptomatic or mild, infections in the elderly are likely to be underreported. The reported yearly incidence in adults is between 4 and 11% and in adults aged over 50 years; hospitalization rates for HMPV were similar to those associated with influenza and RSV [8].

Walsh et al. showed that the risk for symptomatic severe HMPV infection was higher in the elderly. HMPV infection was asymptomatic in 44% of the healthy elderly in contrast to 71% of the healthy young adults. Thirty-eight % of the elderly with HMPV infection used medical care in contrast to 9% in the young adults [23]. Rates for hospitalization in elderly patients over 65 years were also significantly higher for HMPV infection (22.1/10,000 residents) compared to influenza virus (12.3/10,000 residents), but similar to those of RSV infection (25.4/10,000 residents) [8,23]. Antibody levels prior to infection were higher in elderly, suggesting possible immune dysregulation associated with decreased viral clearance in elderly [117].

Several studies have reported outbreaks in long-term care facilities for elderly. Boivin et al. studied a large outbreak in a long term care facility in Canada in which 96 (27%) op the 364 residents had respiratory symptoms. Six out of 13 tested residents were HMPV positive by RT-PCR. Nine patients died, of which three residents tested HMPV positive [103]. In a 23-bed ward in a hospital for older people in Japan, all 8 residents with respiratory symptoms were HMPV positive by RT‑PCR [118]. None of these residents died. Tu et al. found 10 of 13 tested residents of a 53-bed psychiatric ward of an armed forces general hospital in Taiwan HMPV positive by RT-PCR [41]. In a summer outbreak in a long term care facility in California 26 (18%) of residents developed respiratory symptoms. Five of the 13 tested residents were HMPV positive [105]. In an outbreak that the authors of this review described, the attack rate was 13% in a long term care facility [43]. Three patients died, however these were only possible cases. Osbourn et al. found an attack rate of 16.4% in HMPV outbreak in a long‑term care facility in Australia, in which two residents died [104]. Sixteen (36%) of 44 residents in a long-term care facility in Oregon had respiratory symptoms of which 6 of 10 tested residents were HMPV positive by RT-PCR [119]. Another study in a community hospital in England reported an attack rate of 29.4%. The different settings (residential care facilities for elderly versus hospital settings) and different case definitions might partly explain the difference in attack rate and mortality.

Several case reports and case series concerning HMPV infections in immunocompromised patients have been published reporting varying morbidity and mortality [67,120,121,122,123,124]. While immunocompromised patients, including patients with haematological malignancies and solid organ and hematopoietic stem cell transplant (HSCT) patients appear to acquire HMPV infection at the same frequency as immunocompetent individuals, they seem to be at risk for severe infections, probably due to poor viral clearance [125,126,127]. Clinical course is prolonged and respiratory failure may develop [121]. However, Debiaggi showed that HSCT recipients may frequently develop symptomless HMPV infection [128].

Sumino et al. examined a cohort of 688 patients who underwent a bronchoscopy. Of these patients, 72% were immunocompromised (mainly lung transplant patients) and 30% were patients without acute illness who underwent routine bronchoscopy for surveillance after lung transplantation or follow-up of rejection. Six cases of HMPV infection were identified using RT-PCR; four of these were immunocompromised hosts. In the asymptomatic individuals, no cases were identified [129].

Kamboj et al. showed that HMPV is detected in 2.7% of cancer patients with respiratory disease. However, HMPV was associated with mild respiratory disease and RSV and influenza were more often found. In patients with hematologic malignancies HMPV was found more often [58].

Debur et al. showed that HMPV was present in 2.5% of hematologic stem cell transplant recipients with respiratory disease. Most patients presented with upper RTI, while 27% had a lower RTA. No patients died [130].

Englund et al. performed a retrospective survey to demonstrate the importance of HMPV in hematopoietic stem-cell transplant recipients [131]. In 3% of these patients who underwent a BAL because of LRTI was HMPV detected (by RT-PCR). Clinical course in this group was severe and 80% died with acute respiratory failure.

Williams et al. showed that HMPV is found in the same frequency as RSV, influenza and parainfluenzavirus in hematologic malignancy patients with acute respiratory disease. All patients presented with an upper RTI, but 41% progressed to a lower RTI. One third (three patients) of these patients died, however in two of these patients potential bacterial pathogens were also found in their BAL fluid [125].

Cane et al. published a case report about a HSCT recipient who succumbed to progressive respiratory failure following an upper respiratory prodrome and where HMPV was detected as the sole pathogen in the nasopharyngeal aspirate [132].

In lung transplant patients, HMPV was found in 6% of adults with RTI. This was significantly lower then the most frequently found viral cause, namely parainfluenza virus (17%) [133]. RSV and influenza were found in 12% and 14% respectively. The rate of required hospitalization and length of stay of hospitalization were not different between HMPV and other respiratory viruses. In this study, viral RTI was associated with acute graft rejection. However, this rate was significantly higher for RSV infection compared to HMPV infection.

Larcher et al. found HMPV in 25% of BAL fluids from lung transplant patients. Not al of them had respiratory symptoms at the time of the lavage. In this study, HMPV infection seems associated with acute graft rejection, but not with the development of bronchiolitis obliterans [134]. However, other studies suggest that viral RTI is associated with the risk of the development of bronchiolitis obliterans [135,136].