Long-term tolerability and effectiveness of raltegravir in Japanese patients: Results from post-marketing surveillance

Antiretroviral agents are approved in Japan based on non-clinical and clinical data reported from overseas. Neither the long-term tolerability nor the effectiveness of raltegravir or other integrase strand transfer inhibitors in Japan is known. This study reports on the long-term tolerability and effectiveness of raltegravir in Japanese clinical practice using data collected through approximately 9 years of post-marketing surveillance. This observational survey used data on human immunodeficiency virus (HIV) infected patients initiated treatment with raltegravir between 2008 and 2017 in the HIV-related drug (HRD) cooperative survey to assess the safety and effectiveness of raltegravir in real world clinical practice. There were totally 1,303 patients prescribed raltegravir across 30 institutions; 1,293 patients and 1,178 patients were included for the safety and effectiveness analyses, respectively. The overall risk of adverse drug reaction was 17.25%, with abnormal hepatic function and hyperlipidaemia (<1.5%) having the highest proportion. Median HIV-1 RNA viral loads rapidly decreased below 40 copies/mL after 3 months of raltegravir use in treatment-naïve patients, and consistently sustained below 40 copies/mL after the start of raltegravir use in treatment-experienced patients. Among the patients who were treated for 7 years, 92.00% (95% CI: 73.97–99.02) maintained HIV-1 RNA viral load below 50 copies/mL. Additionally, CD4+ cell counts exceeded >500 cells/μL in treatment-naïve and treatment-experienced patients after 3 years and 4 years of treatment, respectively. In Japanese HIV patients, long-term treatment with raltegravir is well-tolerated and effective at viral suppression as measured by HIV-1 RNA levels and subsequent change in CD4+ cell counts. Such benefits can be expected for not only treatment-naïve but also treatment-experienced patients.

The safety and efficacy of raltegravir administration in clinical trials for up to 5 years among HIV-infected patients in North American, South American, Australian, and South Asian countries have been previously reported [1][2][3][4][5][6][7][8]. After its approval in 2007 in the United States, 112 other countries also approved raltegravir [9]. In Japan, use of a 400-mg raltegravir tablet twice daily was approved by the regulatory authority in June 2008, and this drug has been marketed in Japan since July 2008. However, the safety and efficacy of raltegravir has not been evaluated in clinical trials specifically targeting Japanese patients. In Japan, around 1,500 HIV/Acquired immunodeficiency syndrome (AIDS) cases have been reported annually since 2007 and 1,448 HIV/AIDS (1,380 males) cases were reported in 2016. Men who have sex with men (MSM), including bisexual contacts, made up 73% (735/ 1,011) of all HIV cases and the majority was in their 20's to 40's. The cumulative number of reported HIV/AIDS cases (excluding coagulating agent-related cases) was 27,443 as of the end of 2016 [10].
A rapid drug approval process is applied for antiretroviral agents in Japan, and this approval is based on non-clinical and clinical data reported from overseas as an orphan drug. As such, raltegravir was approved for clinical use under the condition of conducting a post-marketing surveillance to collect domestic data on the safety and effectiveness of this agent in clinical practice. Because antiretroviral agents are usually used concomitantly with other anti-HIV drugs, all pharmaceutical companies with an approved anti-HIV drug work together to conduct a HIV-related drug (HRD) cooperative survey [11].
The surveillance of raltegravir through the HRD cooperative survey started at the time of its marketing, in July 2008. To date, no data on the long-term tolerability and effectiveness in Japan of either raltegravir or other INSTI agents are available. Furthermore, only a few studies on the long-term effectiveness of key drug have been conducted in other countries; they include the 12-year experience of nevirapine use in France [12], a 9-year cohort of efavirenz and nevirapine in Senegal [13], a 10-year cohort on efavirenz and nevirapine in Uganda [14], and the 7-year efficacy of a lopinavir/ritonavir-based regimen in the United States [15]. Therefore, the objective of the present study was to assess the long-term tolerability and effectiveness of raltegravir in Japanese clinical practice, using the HRD cooperative survey data collected as post-marketing surveillance over the approximately 9 years from 2008 to 2017.

Study design
This observational survey used post-marketing surveillance data collected for the HRD cooperative survey and was conducted in accordance with the requirements of the pharmaceutical affairs law and the ministerial ordinance of 'Good Post-Marketing Study Practice (GPSP)'. At their discretion, investigators of the contracted institutions registered patients who were newly prescribed with particular antiretroviral agents and provided data on the safety and effectiveness of the antiretroviral agents for each treated patient. Patient background information, such as age, sex, comorbidity on administration (including hepatic disease and renal disease), presence of allergy, status of anti-HIV drug use, and concomitant drug use (dose level, dosing period), was collected at registration. Data were collected by filling out a case report form upon registration and annually thereafter. If a patient experienced an adverse event (AE), the corresponding data were reported immediately using the AE report form. The management of the HRD cooperative survey was delegated to CMIC PMS Co., Ltd. (Tokyo, Japan). All data were fully anonymized before the authors of this study accessed them.

Study population
Data on patients registered as newly treated with raltegravir between July 2008 (the beginning of raltegravir marketing) and March 2017 were drawn from the HRD cooperative survey for this study. Of the 33 institutions in which patients were prescribed raltegravir, three institutions did not agree to the disclosure of information, so they were excluded from these analyses.

Safety assessments
To assess the safety of raltegravir, AEs and abnormal variations in laboratory test parameters (i.e. haematology test, clinical biochemical test, urine analysis, and ocular abnormalities) were collected during the treatment period. All events with a relationship between the AE and drug were defined as adverse drug reactions (ADRs). AEs and ADRs were classified according to the Medical Dictionary for Regulatory Activities/Japanese version (MedDRA/J Ver.20.0).

Effectiveness assessments
For the assessment of effectiveness, the CD4 + cell counts, HIV-1 RNA levels, and HIV infection clinical stages (i.e. Centers for Disease Control and Prevention [CDC] categories [16]) before and after raltegravir administration (baseline and treatment period) were collected. Infection clinical stages were classified as follows: category A was defined as asymptomatic HIV infection, category B was defined as symptomatic HIV infection, and category C was defined as acquired immunodeficiency syndrome (AIDS).

Statistical analysis
Patient demographics, clinical characteristics, AEs, and ADRs were summarized descriptively from the collected data. The long-term safety of raltegravir was assessed by summarizing the risk of ADRs by the time of occurrence from the start of raltegravir treatment: <90 days, between 90 and <180 days, between 180 and <270 days, between 270 days and <1 year, between 1 year and <2 years, between 2 and <3 years, between 3 and <4 years, between 4 and <5 years, between 5 and <6 years, and �6 years. To assess potential risk factors for ADRs, the ADR occurrence was further analysed according to baseline characteristics using a Fisher's exact test or Cochran-Armitage test. Changes in the median HIV-1 RNA levels and median CD4 + cell counts were calculated over time at 0, 3, 6, 12, 24, 36, 48, 60, 72, 84, and 96 months after the start of raltegravir treatment for treatment-naïve patients and treatmentexperienced patients. The proportion of patients with <50 copies/mL of HIV-1 RNA, including blip cases (i.e. patients with <50 copies/mL of HIV-1 RNA with transient increases up to 1,000 copies/mL), was also calculated for the virologic assessment, and its 95% confidential intervals was estimated using Clopper-Pearson method. All statistical tests were two-sided at a significance level of 0.05 and were conducted using SAS Release 9.4 (SAS Institute, Inc., Cary, NC, USA).

Patient disposition
A total of 1,303 patients were prescribed with raltegravir across 30 institutions of the HRD cooperative survey. Of these, 10 patients were excluded due to the fact that their prescription of raltegravir was written before its launch in Japan. 1,293 patients were included for the safety analysis. For the analysis of raltegravir effectiveness, one patient with an off-label use of raltegravir and 115 patients with no recorded CD4 + cell counts, number of HIV-1 RNA copies, or CDC categories either before or after raltegravir administration were excluded. Finally, data from the remaining 1,178 patients were included for the effectiveness analysis (Fig 1).

Patient demographics and clinical characteristics
The patient demographics and clinical characteristics at baseline are summarized in Table 1.
In the safety analysis set (n = 1,293), 94.1% of patients were male, and the median age at baseline was 41 years (Interquartile range, IQR: 34-50 years). The majority of patients belonged to pre-treatment CDC category A (54.5%), followed by categories C (30.9%) and B (6.0%). (n = 55; 4.3%), and hyperuricaemia (n = 50; 3.9%). Most patients (99.1%) were prescribed with a dose of 800 mg of raltegravir per day. The median treatment period from the start of raltegravir administration was 1,123 days, and 84.2% of the patients had been treated with raltegravir longer than 1 year.

Analysis for effectiveness parameters
Virologic responses. At the start of the treatment, the median HIV-1 RNA viral loads among the treatment-naïve patients were considerably higher compared with those of treatment-experienced patients (64,000 vs. 39 copies/mL), but this level rapidly decreased to 78.5 copies/mL after one month of raltegravir treatment. Median HIV-1 RNA viral loads in treatment-naïve and treatment-experienced groups each maintained below 40 copies/mL after three months of raltegravir treatment (Fig 2, S2 Appendix). Regarding long-term virologic suppression in both treatment-naïve and treatment-experienced patients, the proportion of patients with HIV-1 RNA viral loads of less than 50 copies/mL (including blip cases) was 81.79% (503/615 cases, 95%CI: 78.51-84.76) at 3 months and 89.47% (620/693, 86.94-91.65) at  Immunologic responses. At the start of raltegravir treatment, the median CD4 + cell counts in treatment-naïve patients were lower compared with those in treatment-experienced patients (216 vs. 408 cells/μL), but the CD4 + cell counts in the treatment-naïve group reached 346 cells/μL after 3 months and 522 cells/μL after 36 months of raltegravir treatment (Fig 4,  S4 Appendix). The CD4 + cell counts in both the treatment-naïve and treatment-experienced groups displayed an increasing trend throughout the treatment period; the CD4 + cell counts were maintained at over 500 cells/μL after 36 months in treatment-naïve patients and after 48 months in treatment-experienced patients.
CDC categories. Most of the patients assessed as belonging to category A (505/512 cases; 98.6%) and category B (51/53 cases; 96.2%) maintained their pre-treatment CDC category even after raltegravir administration. Only nine cases had disease progression.

Discussion
This study was conducted using the results of a nine-year post-marketing surveillance on raltegravir in Japan. Data from 1,303 patients were initially extracted from the HRD cooperative survey, and, after excluding 10 ineligible patients, a total of 1,293 patients were included in the safety analysis. The overall ADR risk (17.25%) in this survey was fairly low, compared with previous international clinical trials reporting the ADR incidence for raltegravir, which ranged from approximately 44%-61% with study periods spanning 1-5 years [1][2][3][4][5][6][7][8]. Hyperlipidaemia and abnormal hepatic function were the most commonly reported ADRs in our study, with risks slightly exceeding 1%. No specific tendency was observed in the occurrence of other ADRs, including rash, cardiovascular disorders, elevated creatine kinase, and musculoskeletal and connective tissue disorders (i.e. rhabdomyolysis and myopathy) reported in previous clinical trials with a 2-year maximum of raltegravir administration [1,2,4,17]. Muscle symptoms including myalgia, pain in extremity and rhabdomyolysis (0.31%, 4 cases) and CPK increase (0.15%, 2 cases) were fairly low in this survey, compared with other observational study (SCOLTA project) in patients receiving raltegravir, which were 5.2% and 21.1%, respectively  Tolerability & effectiveness of raltegravir [18]. SCOLTA project also investigated drug-related central nervous system (CNS) symptoms which were 10.4%, whereas all psychiatric and nervous system disorders in this survey were low at 2.01% (26 cases) [19]. As for neoplasms, either benign or malignant, Kaposi's sarcoma could be considered a disease-specific complication, but the proportion of all remaining malignancies was as low as 0.08%. For long-term safety, no specific tendency was observed in the overall ADR risk by raltegravir treatment period, but the risk of metabolism and nutrition disorders increased as the raltegravir treatment period lengthened. However, nearly 40% of the patients were 45 years of age or older at the initiation of raltegravir and many patients have comorbidities including diabetes mellitus or hyperlipidaemia, therefore it cannot be concluded that the trend was caused by long-term use of raltegravir. Although the risk of malignancies in patients treated with raltegravir should continue to be monitored, the <1% risk of most reported ADRs in this long-term safety assessment of a large Japanese cohort with over 1,000 patients suggests the general tolerability of raltegravir.
The effectiveness analysis was conducted on data from a total of 1,178 patients. The median HIV-1 RNA viral loads in treatment-naïve patients decreased rapidly after one month of raltegravir treatment and were identical in both treatment-naïve and treatment-experienced patients after 3 months of raltegravir treatment. Although decreasing numbers on raltegravir treatment as time goes on, using this as denominator, the proportion of patients with <50 copies/mL of HIV-1 RNA remained at 92.00% (23/25 cases, 95%CI: 73.97-99.02) in patients treated for 7 years, which is higher than or at a similar level as the corresponding proportions reported by previous clinical trials conducted overseas for a maximum of 5 years [1][2][3][4][5][6][7][8], suggesting a long-term virologic suppression by raltegravir treatment in Japanese patients.
Similarly, the median CD4 + cell counts in treatment-naïve patients were initially lower than those in treatment-experienced patients, but they rapidly increased to 346 cells/μL after 3 months of raltegravir treatment and were maintained at >500 cells/μL after 3 years of raltegravir in treatment-naïve patients and after 4 years in treatment-experienced patients. Previous studies reported that the changes in CD4 + cell counts from baseline were 189 cells/μL, 240 cells/μL, and 374 cells/μL in treatment-naïve patients (STARTMRK study) and 109 cells/μL, 123 cells/μL, and 293 cells/μL in patients with triple-class drug resistance in whom previous ART had failed (BENCHMRK-1 and -2 study) after approximately 1 year, 2 years, and 5 years of raltegravir administration, respectively [3][4][5][6][7][8]. In the present survey, we found an increase by 218 cells/μL, 278 cells/μL, and 325 cells/μL in treatment-naïve patients and by 41 cells/μL, 72 cells/μL, and 127.5 cells/μL in treatment-experienced patients after 1-, 2-, and 5-year treatment with raltegravir, respectively; this result may indicate an equivalent or better improvement regarding CD4 + cell counts resulted from virological suppression in raltegravir-treated Japanese patients. The observed improvement in the immunologic response even for treatment-experienced patients implies that an improvement for patients is likely after changing their ART regimen to include raltegravir.
There are several limitations to the present study. First, our results may be specific to particular groups of patients with HIV infection. The included patients in this study were registered at designated institutions, and, although in principle all patients receiving raltegravir were registered, each enrolment decision was made by contractor investigators. Therefore, a selection bias is inevitable. Second, the effectiveness results in this study cannot be purely attributed to raltegravir because antiretrovirals are commonly used in combinations. The present results support the effectiveness of raltegravir, but they must be interpreted with care. Third, as this data was drawn from HRD cooperative survey conducted by all companies having an HIV drug approved in Japan, we can only extract data of patients still on treatment with raltegravir at the time of the annual report. If patients collected as raltegravir data and switched for viral loads �50 copies/mL judged from collected effectiveness data, these patients were counted as failure. However, after they have switched to another ART regimen, we might not have been able to follow them any longer in this study. Therefore, a selection bias for over estimating effectiveness is inevitable. Fourth, as we had no control group for comparison, the safety and effectiveness results for raltegravir cannot be compared with those in HIV-infected patients who received no or other antiretroviral treatments. Last, data on patients who were treated with raltegravir for longer than 7 years were scarce, which may have limited the precision of those results.
In conclusion, long-term treatment with raltegravir is well-tolerated and effective regarding viral suppression, as measured by HIV-1 RNA levels and CD4 + cell counts, in Japanese HIV patients. Such benefits can be expected not only for patients who are about to start antiretroviral therapy but also for patients who switch antiretroviral regimens to one including raltegravir.