Introduction

Yearly influenza epidemics have raised medical and social concerns because they are associated with considerable morbidity and mortality [1]. Two countermeasures, vaccination and antivirals, are available to control human influenza infections [2]. Although vaccinations may play a critical role in influenza prophylaxis, they are insufficient in a pandemic because of the time required to produce a new vaccine. Therefore, antivirals are an important tool to mitigate the effects of an influenza pandemic. Neuraminidase inhibitors (NAIs), rather than M2 inhibitors, have been preferred for the treatment of influenza infection, owing to the continuing high prevalence of influenza A virus subtype H1N1pdm ((H1N1)pdm09) and influenza A virus subtype H3N2 (A(H3N2)) resistance to the latter drugs [3, 4]. In this situation, the licensing of new NAIs other than oseltamivir and zanamivir has important implications as an alternative option for the treatment of influenza infection [5]. In 2010, two new NAIs, laninamivir and peramivir, were approved for clinical use for the treatment of influenza A and B infection in Japan, as one of the earliest countries to license these compounds. Clinicians have limited experience of using these drugs other than information from clinical trials [610]. In the present study, we report on the effectiveness of NAIs, including early data for the two newly released drugs, peramivir and laninamivir, by observing fever alleviation in influenza-infected patients in the 2010-2011 season.

Patients and methods

Study procedures

This study was conducted in the 2010-2011 winter season from December 27, 2010, to March 29, 2011, in Japan. Two outpatient clinics in two prefectures, Niigata and Gunma, participated in this study. As a rule, randomly selected patients with influenza-like-illness (ILI) who visited the clinics were enrolled. ILI was defined as sudden onset of fever (≥37.5 °C) and respiratory symptoms, as well as headache, arthralgia, or myalgia. Clinicians used influenza rapid diagnostic test kits licensed in Japan in the patients who had ILI symptoms to screen for influenza A or B infection. Of the commercially available antigen detection kits based on immunochromatography, Quick Ex flu (Denka Seiken, Tokyo, Japan), ESPLINE Influenza A&B-N (Fujirebio, Tokyo, Japan), and Check FluA·B (Alfresa Pharma, Osaka, Japan) were mainly used. If the rapid test was positive, patients were given one of four NAIs, oseltamivir, zanamivir, laninamivir, or peramivir, according to the advice of clinicians and the preference of the patients or their guardians. Upon enrollment in this study, written informed consent was obtained from the patient or their guardian for the collection of clinical information and specimens for virological investigations.

Oseltamivir was prescribed to be taken orally twice daily for 5 days (150 mg per day for adults and for children who weighed ≥37.5 kg, and 4 mg/kg for children who weighed <37.5 kg), and zanamivir inhalation was prescribed to be used twice daily for 5 days (20 mg per day for adults and for children ≥5 years old). Laninamivir was prescribed as a single inhalation (40 mg for adults and for children ≥10 years old, or 20 mg for children <10 years old as a treatment course), and peramivir was administered as a single intravenous drip infusion at the outpatient clinic (300 mg for an adult without complications, 600 mg for adult patients at high risk of developing complications, and 10 mg/kg for children, with a maximum dose of 600 mg as a treatment course). All dosages followed the advisories issued by the Ministry of Health, Labor, and Welfare of Japan. Single use of antipyretics was allowed when a patient had a fever of ≥38.5 °C. Clinicians were permitted to prescribe other drugs such as cough suppressants and expectorants depending on the patient’s complaints.

Age, sex, vaccination status, body weight, and body temperature were recorded for all eligible patients. The date and time of onset of the fever, and the dates of clinic visits were recorded by the doctors. Time of onset was defined when a patient had a fever of more than 37.5 °C before the clinic visit. Time of administration of the first dose of antiviral was defined as the time of the clinic visit, because clinicians administered the drug at the clinics (peramivir), or instructed the patients to take the drug as soon as they reached home (oseltamivir and zanamivir) or at the pharmacy (laninamivir). Patients were asked to measure and record axillary body temperature at least three times per day (9 a.m., 12 p.m., and 8 p.m.) at home. Then they sent the record of information on daily body temperature by mail, or brought the record directly to the clinic after alleviation of the fever. The time at which the body temperature dropped below 37.5 °C without recurrence was defined as the time at which the patient became afebrile. This study was approved by the Medical Faculty Ethics Committee of the Niigata University Graduate School of Medical and Dental Sciences.

Virological investigation

Specimens from throat swabs, nasal swabs, or nasal aspirates taken at the time of the clinic visits were submitted for virus isolation.

Clinical samples were inoculated on Madin-Darby canine kidney (MDCK) cells for virus isolation. The typing and subtyping of influenza virus isolates into A(H1N1)pdm09, A(H3N2), and type B was performed by hemagglutinin inhibition test or a real-time polymerase chain reaction (PCR) method [11]. Analysis of oseltamivir resistance in A(H1N1)pdm09 samples was performed by using real-time PCR with a cycling probe assay which can specifically detect the H275Y substitution in the NA gene conferring oseltamivir resistance [12].

Statistical analysis

For estimating fever duration after administration of the first dose of the antiviral in the patients who received oseltamivir, zanamivir, laninamivir, or peramivir, Kaplan–Meier analyses were applied and the log rank test was used for comparisons of estimated fever duration among the treatment groups. For multivariate analysis, we used a Cox proportional-hazards model to evaluate the duration of fever from the first dose of the NAI to fever alleviation <37.5 °C. We set the duration of fever (hours) as a dependent variable, and set the following factors as independent variables: the type of treatment (oseltamivir, zanamivir, laninamivir, or peramivir), age, sex, body weight, vaccination status, time from onset to the clinic visit, and body temperature at the clinic visit. All statistical analyses were performed using SPSS software, version 19.0 (IBM, Armonk, NY, USA). P values of <0.05 were employed to define statistical significance.

Results

Patient characteristics

A total of 681 patients in the two clinics were enrolled in the study during the 2010–2011 season. After excluding patients for whom there was incomplete clinical information, data for analysis were obtained from 263 eligible patients; that is, 211 with A(H3N2) virus infection, 45 with A(H1N1)pdm09 infection, and 7 with influenza virus type B infection. The demographic characteristics of the patients are shown in Table 1. Of the 211 patients infected with A(H3N2) virus, 104 were treated with oseltamivir, 94 were treated with zanamivir, 9 were treated with laninamivir, and 4 were treated with peramivir. Of the 45 patients infected with A(H1N1)pdm09 virus, 15 were treated with oseltamivir and 30 were treated with zanamivir. None of the patients who were infected with A(H1N1)pdm09 virus were treated with laninamivir or peramivir. Further evaluation was done only for patients infected with the A(H3N2) and A(H1N1)pdm09 viruses, because the sample size for the influenza B virus-infected patients was too small (N = 7) to analyze in detail. In the patients infected with A(H3N2) virus, the average age in the oseltamivir treatment group was significantly lower than that in the other treatment groups. The time from the onset of fever to the first clinic visit was shorter in the peramivir treatment group than in the other groups. The average body temperature during the clinic visit was significantly higher in the oseltamivir group than that in the zanamivir group.

Table 1 Baseline demographic and clinical characteristics of patients infected with influenza A virus subtype H3N2 and influenza A virus subtype H1N1pdm in the 2010–2011 season

Regarding the patients infected with A(H1N1)pdm09 virus, the average age in the oseltamivir treatment group was significantly lower than that in the zanamivir treatment group. In both the A(H3N2) and A(H1N1)pdm09 virus-infected patients, no other statistically significant differences in baseline status items such as sex, time to the first clinic visit, or vaccination status were detected among the patients who received oseltamivir, zanamivir, laninamivir, and peramivir (Table 1).

In the genetic analysis, no H275Y substitution in the NA gene conferring oseltamivir resistance was detected in any of the 45 A(H1N1)pdm09 strains.

Duration of fever after administration of the first dose of antiviral therapy

For the A(H3N2)-infected patients, the duration of fever was evaluated in the four treatment groups–oseltamivir, zanamivir, laninamivir, and peramivir. Kaplan–Meier analysis showed that the time to alleviation of fever in the peramivir treatment group (17.0 h, 95 % CI 7.2–26.8 h) was significantly shorter than that in the oseltamivir treatment group (P = 0.044, Fig. 1). No significant difference was found in times to fever alleviation among the other three treatment groups–oseltamivir, zanamivir, and laninamivir.

Fig. 1
figure 1

Kaplan–Meier curves of time to return to body temperature <37.5 °C in the influenza A(H3N2)-infected patients (a), and the influenza A(H1N1)pdm09-infected patients (b)

In the influenza A(H1N1)pdm09-infected patients, only oseltamivir and zanamivir treatment groups were compared, because there were no groups treated with the new drugs. In the Kaplan–Meier analysis, no significant difference was detected between the oseltamivir and zanamivir treatment groups in the time to alleviation of fever (Fig. 1).

In the A(H3N2)-infected patients, the Cox proportional-hazards model showed that the time to alleviation of fever was 3.3 times shorter in the peramivir treatment group than that in the oseltamivir treatment group, with an adjusted P value of 0.0303 (Table 2). No significant difference in the time to alleviation of fever was found in the zanamivir and laninamivir treatment groups when compared with the oseltamivir treatment group. In the A(H1N1)pdm09-infected patients, the Cox proportional-hazards model showed no statistically significant difference in the time to alleviation of fever between the oseltamivir and zanamivir treatment groups.

Table 2 Time to alleviation of fever analyzed using the Cox proportional-hazards model

Discussion

In this study, we evaluated the clinical effectiveness of NAIs by analyzing the duration of fever after initiation of therapy in patients with influenza A infection treated with oseltamivir, zanamivir, laninamivir, or peramivir in Japan. We found no statistically significant difference between the oseltamivir treatment group and the zanamivir treatment group in the patients who were infected with either the A(H3N2) or A(H1N1)pdm09 viruses. Kawai et al. reported that oseltamivir was more effective than zanamivir in influenza A(H3N2)- or A(H1N1)pdm09-infected patients who were mostly adults [13, 14]. On the other hand, Sugaya et al. [15] showed that there was no significant difference in the effectiveness of oseltamivir and zanamivir against A(H3N2) in children. Our result may support the study of Sugaya et al. because most of the patients in our study were children.

We found that A(H3N2)-infected patients who underwent peramivir therapy showed faster alleviation of fever compared to that shown in A(H3N2)-infected patients treated with the other three antivirals. Even after adjustment by the factors age, sex, body weight, vaccination status, time from fever onset to the clinic visit, and body temperature at the clinic visit, patients who received peramivir still showed significantly faster reduction of fever compared with the findings in the other study groups.

In a previous study, Kohno et al. [6] showed the efficacy of peramivir therapy in adult patients with seasonal influenza virus infection (A(H1N1) and A(H3N2)) compared with results in a placebo group, and in a randomized control study [7], they demonstrated the non-inferiority of peramivir, compared with oseltamivir, in reducing the time to alleviate influenza symptoms in patients with seasonal influenza virus infection (A(H1N1)pdm09, A(H3N2), or influenza virus type B). Sugaya et al. [8] reported the efficacy of peramivir treatment in A(H1N1)pdm09-infected children. However, so far only a few studies have documented the efficacy of peramivir in patients infected with influenza virus; thus, the present study is still meaningful from the clinical point of view. In addition, at present, only in Japan peramivir has been approved for clinical use. Thus, it is important to document early experiences in its clinical use in Japan, as shown in our study.

In our study, the sample size for analysis was small, especially in the peramivir and laninamivir treatment groups. This limitation may affect the reliability of the study. Also, in the peramivir group, the time from fever onset to the clinic visit was shorter than that in the other groups. Perhaps the early visit to the clinic in the peramivir treatment group may have led to the shorter duration of fever. However, the duration of fever in the peramivir group was still significantly shorter than that in the other study groups when we compared fever duration using multivariate analysis adjusting for factors including time from fever onset to the clinic visit.

Another limitation of our study is that the antiviral drug prescribed depended on the discretion of clinicians or the preference of patients or their guardians. Laninamivir and peramivir were newly released in Japan in 2010. In addition, in most cases, oseltamivir was not advised for use by patients of 10–19 years of age in Japan because of reports of an association with neuropsychiatric symptoms in this age group [5]. Zanamivir is not recommended for patients with underlying respiratory disease or for children <5 years old. Therefore, the decision whether to administer one of the four drugs to patients was left mostly to the discretion of clinicians, who followed the Japanese guidelines outlined above and patient preference. This non-randomized manner of drug allocation may have caused selection bias in this observational study.

Peramivir is an anti-influenza drug that selectively inhibits the neuraminidase of human type A and type B influenza viruses. The most important characteristic of peramivir is its rapid bioavailability when administered intravenously. Current antiviral treatments other than peramivir– such as oseltamivir, zanamivir, and laninamivir–are administered either orally or by inhalation. These routes may not provide rapid, reliable drug delivery in seriously ill patients. In view of the certainty of drug delivery with intravenous administration, peramivir may be administered as the first-line therapy, especially in patients who have high-risk factors for complications or those who cannot take drugs orally or inhale drugs.

In our study, in the patients with influenza A(H3N2) infection, no significant difference in duration of fever was shown in the laninamivir treatment group compared with the other study groups. Watanabe et al. [10] showed the non-inferiority of laninamivir, compared with oseltamivir, in patients with seasonal influenza (A(H1N1)pdm09 and A(H3N2)). Sugaya and Ohashi [9] showed more rapid alleviation of influenza illness in children infected with seasonal A(H1N1) who received laninamivir than in those who received oseltamivir. The reason for the slower alleviation of influenza illness in patients with seasonal A(H1N1) virus infection who received oseltamivir than in those who received laninamivir was possibly that almost all seasonal A(H1N1) viruses possessed the H275Y mutation, which conferred resistance to oseltamivir [16].

Inhaled laninamivir was developed in Japan and approved for use in our country in 2010. Laninamivir octanoate has been shown to have neuraminidase inhibitory activity against various influenza A and B viruses, including oseltamivir-resistant viruses [10]. The chemical structure of the active drug, laninamivir, is similar to that of zanamivir. The most important characteristic of laninamivir octanoate is its long-lasting antiviral activity, and because of this, laninamivir is administered in the form of a single inhalation on the first day of treatment, and it stays active in the respiratory tract for several days [17].

During the 2009 influenza pandemic, the use of oseltamivir increased dramatically, and this raised a concern about the emergence of oseltamivir resistance that could limit the effectiveness of NAIs. Fortunately, oseltamivir-resistant A(H1N1)pdm09 strains are still rare in the community [18, 19]. However, sporadic cases of oseltamivir-resistant A(H1N1)pdm09 infections were detected in Japan soon after community circulation of the A(H1N1)pdm09 virus was observed [20]; thus, the spread of drug-resistant viruses has become a major concern in the post-pandemic period [16, 21, 22]. We must always pay attention to appearance of resistant viruses possessing the H275Y mutation conferring oseltamivir and peramivir resistance that may reduce the clinical effectiveness of these drugs.

In this report, we demonstrated the clinical effectiveness of the new NAIs. Because the number of cases we studied was limited, further clinical study to investigate the effectiveness of new NAIs in influenza infections is important.