Health-Related Quality of Life With Enfuvirtide (ENF; T-20) in Combination With an Optimized Background Regimen

Summary: This study assessed the impact of enfuvirtide on health-related quality of life (HRQoL). Patients enrolled in 2 phase 3 trials T-20 versus Optimized Regimen Only (TORO 1 and 2) completed the Medical Outcomes Study (MOS)-HIV questionnaire at baseline and at 4, 8, 16, and 24 weeks. A total of 995 treatment-experienced HIV-1-infected individuals received either self-administered enfuvirtide (90 mg twice daily) + optimized background (OB) or OB alone and had at least 1 follow-up visit. Data from the 2 clinical trials were pooled. Analysis of covariance was used to evaluate changes in the 10 MOS-HIV scale scores and 2 summary scores. Least-squares means for these changes were calculated and used to test for between-group differences. There were no significant between-group differences in any HRQoL measure at baseline. Most MOS-HIV scores showed improvement in the enfuvirtide arm compared with OB alone, although only some of these were significant. Improvements in the general health scale were significantly higher in the enfuvirtide arm compared with OB alone at all post-baseline time points. No scale or summary score for the OB arm showed a significantly greater improvement in score from baseline compared with the enfuvirtide arm, at any time point. The mental health summary score at 24 weeks was significantly higher in the enfuvirtide arm compared with OB alone. Enfuvirtide in addition to an OB regimen does not adversely affect and may improve HRQoL when self-administered for up to 24 weeks by treatment-experienced, HIV-1-infected individuals.

Since the introduction of highly active antiretroviral (ARV) therapy, there has been an unprecedented decrease in the mortality and morbidity associated with AIDS.^1,2 Despite this, the durability of viral suppression is often limited. There are numerous reasons for failure of ARV regimens, including limited antiviral potency of existing therapeutic regimens, toxicity, alterations in bioavailability and metabolism of certain therapeutic agents, and lack of adherence to complex treatment regimens. The result is often virologic failure, which, in the face of ongoing therapy, leads to the emergence of viral resistance to one or more of the ARV treatments.

Initial emergence of resistance to one member of a class may be associated with cross-resistance to other members of that same class. Furthermore, transmission of drug-resistant virus is increasing.³ Thus a new ARV with a different mode of action that works at a different stage of the virus replication cycle would make a valuable addition to the currently available drugs.

Enfuvirtide is the first of a new class of ARVs called the HIV-1 fusion inhibitors that work by blocking the fusion of HIV-1 to the host cell membrane, thus preventing the virus from entering the cell. Enfuvirtide has been shown to be efficacious when administered in combination with optimized background (OB) therapy to treatment-experienced, HIV-1-infected individuals.^4,5 After 48 weeks of therapy, pooled data from 2 large pivotal phase 3 clinical trials, T-20 versus Optimized Regimen Only (TORO 1 and TORO 2), showed that the addition of enfuvirtide to an OB regimen resulted in a significantly greater reduction in plasma viral load compared with an OB regimen alone, with a difference of -0.85 log₁₀ copies/mL (P < 0.0001).⁶ The addition of enfuvirtide to an OB regimen also resulted in a significantly greater increase in the CD4⁺ cell count at 48 weeks compared with the OB regimen alone, with a difference of 46 cells/mm³ (P < 0.0001).

Unlike other ARV agents that are administered orally, enfuvirtide is a peptide that would be expected to be rapidly catabolized in the gastrointestinal tract and therefore must be administered by subcutaneous injection. As enfuvirtide works extracellularly and would not be expected to significantly penetrate into cells, it is expected to have a good toxicity profile. This has been confirmed by safety data up to week 48 from TORO 1 and TORO 2.^4,5,7 In TORO 1 and TORO 2, the safety profile of enfuvirtide + OB was shown to be generally comparable to that of OB therapy alone. However, administration of enfuvirtide was associated with a high incidence of injection-site reactions (ISRs), which occurred in most patients self-injecting enfuvirtide (98%). Most ISRs were mild and over 48 weeks, few patients (4.4%) discontinued enfuvirtide therapy as a result.⁷ Over 48 weeks, 11% of patients had ISRs requiring analgesic therapy or limiting usual activities.⁷ The proportions of patients reporting at least one treatment-related adverse event in the first 48 weeks of the TORO studies were 97.1 and 90.7% in the enfuvirtide + OB and OB arms, respectively.⁷

Both HIV disease and HIV therapy can affect health-related quality of life (HRQoL). Improvements in surrogate markers of progressive HIV disease, including HIV RNA levels and CD4⁺ cell counts, have been found to be significantly associated with improvements in HRQoL.^8,9 Although ARV therapy can reduce HIV RNA, increase CD4⁺ cell count, and delay HIV disease progression, the benefits of treatment with some ARV agents may be offset by a decrease in HRQoL attributable to adverse effects. In turn, a treatment-related decline in HRQoL during therapy may lead to discontinuation.¹⁰ The impact of ARV therapy on HRQoL is therefore an important consideration in the selection of treatment for HIV-infected individuals.

Study Objective

The objective of this study was to assess the impact of enfuvirtide, self-administered twice daily (bid) by subcutaneous injection, on HRQoL by patients participating in TORO 1 and TORO 2 for up to 24 weeks. This was achieved by comparing the HRQoL domains, derived from the Medical Outcomes Study (MOS)-HIV, in recipients of enfuvirtide + OB therapy with those in patients given OB alone at 4, 8, 16, and 24 weeks.

METHODS

The MOS-HIV is a multidimensional, disease-specific measure of HRQoL that has been used extensively in studies of HIV-infected patients receiving ARV therapy.^11-16 The questionnaire includes 10 dimensions (physical function, pain, role, social and cognitive functioning, mental health, vitality, health distress, health perceptions, and QoL) and 2 summary scores (physical health and mental health) derived from factor analysis. The 10 dimensions have been found to be both reliable and valid in 6 separate studies.^17-22 The reliability and validity of the 2 summary scores was examined separately by Revicki et al²³ and the summary scores were found to be reproducible across different samples of HIV/AIDS patients and to be reliable and valid measures for demonstrating treatment impact on patient functioning and well-being. The summary scores and the subscales have acceptable internal consistency, test-retest reliability and validity, and are able to detect clinically important differences.¹⁸

The sensitivity of MOS-HIV scores has also been demonstrated. In 3 studies, improved MOS-HIV scores have been shown to correlate with increased CD4⁺ cell count, lower viral load, and fewer HIV-related symptoms¹¹ as well as with patient continuation of therapy.^8,11,17

The MOS-HIV instrument was used in this study to assess the relationship between HRQoL and treatment with enfuvirtide.

Study Design

TORO 1 and TORO 2 are open-label, randomized, phase 3, multinational, controlled clinical trials comparing the efficacy and safety of self-administered enfuvirtide (90 mg bid) in combination with an OB regimen with that of OB alone.^4,5

Patients were asked to complete the MOS-HIV at baseline and at their scheduled 4-, 8-, 16-, and 24-week visits. The questionnaire was given before any clinical procedures or consultations to ensure that patients responded independently of the clinical evaluation.

As TORO 1 and TORO 2 are multinational clinical trials, involving patients from several countries, 14 different language versions of the MOS-HIV were used as appropriate: Danish, Dutch, English (UK, US, and Australian), French (European and Canadian), German, Italian, Portuguese (European and Brazilian), Spanish (European and US), and Swedish. All language versions were validated previously and had been shown to have similar psychometric properties (reliability and validity) as the original version of the MOS-HIV in US English.^24-26

Scores for each of the 10 MOS-HIV scales and both summary scores, measured at 4, 8, 16, and 24 weeks, were totaled and converted to a 0- to 100-point scale, with 100 indicating better health status.^23,27 The methodology for calculating the summary scores, reflecting the physical health and mental health components, is described elsewhere.^23,27

Patients

A total of 1013 patients were enrolled and randomized in either TORO 1 or TORO 2. For the efficacy analysis, the intent-to-treat (ITT) population was used, consisting of all patients who were randomized, received 1 dose of study medication, and had 1 posttreatment HIV-1 RNA level obtained. Eighteen patients were excluded from this group; thus the ITT population consisted of 995 patients (661 in the enfuvirtide + OB arm and 334 in the OB arm).

The patient selection criteria for the 2 trials were similar. All patients were HIV-1-infected adults or adolescents (aged ≥16 years) with HIV-1 RNA ≥5000 copies/mL and ≥3 months' (TORO 2) or ≥6 months' (TORO 1) experience and/or documented resistance to each of the 3 currently licensed classes of ARVs. Patients who had previously received enfuvirtide, or who were receiving other medications, including ARVs, immunomodulators, biologic response modifiers, and chemotherapy agents that were not part of their OB regimen were excluded. Pregnancy or alcohol/substance abuse resulting in an inability to adhere to the protocol, or documented noncompliance to previous ARV treatment regimens, also resulted in exclusion from these clinical trials.

All patients gave written informed consent, and, prior to initiation of the study, the independent ethics committee or institutional review board for each of the centers involved in the study approved the research protocols.

Data Analysis

Data from the 2 clinical trials were pooled for analysis. Combining the data was appropriate because the 2 trials were conducted using very similar clinical research protocols and enrollment criteria and had similar baseline characteristics.^4,5

Data analysis was performed using the ITT population. In the case of patients who withdrew from the study or experienced virologic failure (as defined in the clinical trial protocol), MOS-HIV scores were censored at the time of withdrawal or failure and missing values imputed using the last observation carried forward method.

Analysis of covariance (ANCOVA) was used to evaluate changes in the HRQoL endpoints. The ANCOVA models were specified a priori. The dependent variable was the change from baseline HRQoL score, calculated for each patient as the difference between the MOS-HIV scale score at weeks 4, 8, 16, and 24 and the patient's baseline MOS-HIV score. The model included terms for treatment group (enfuvirtide + OB or OB), MOS-HIV score at baseline, stratum (a classification variable), and the treatment-stratum interaction. Stratum consisted of 4 categories, based on the HIV-1 RNA level at screening visit 2 (<40,000 or ≥40,000 copies/mL) and whether or not newly approved/investigational ARV agents were part of the patient's OB treatment regimen at that time. The treatment-stratum interaction consisted of classifying each patient into 1 of the 4 possible HIV-1 RNA level/ARV combinations.

A separate ANCOVA model was estimated for each MOS-HIV scale and for both summary scores, yielding a total of 12 model estimates for each of the 4 time periods. Overall F-tests were used to assess the overall significance of the models. Least-squares means comparing the predicted MOS-HIV scores for the enfuvirtide + OB treatment group with those for the OB treatment group were also calculated from these models. To determine the direction, size, and statistical significance of the treatment effect, t tests assessing the significance of least-squares means differences were used.

RESULTS

Demographics

The baseline demographic and clinical characteristics of all patients in TORO 1 and TORO 2 are shown by treatment group at randomization in Table 1. All baseline demographic and clinical characteristics were balanced across the treatment groups.

Table 1

Of those patients who remained in the study, response rates to the MOS-HIV were 95.2, 96.8, 97.0, and 97.0% at weeks 4, 8, 16, and 24, respectively. There was no significant difference in response between the enfuvirtide + OB and OB treatment arms at any time point. However, response rates for the individual scales on the instrument varied from 89.9% for the general health scale in the OB group at week 24 to 96.4% in the physical function scale in the enfuvirtide + OB group at week 24.

There were no significant between-group differences in any HRQoL measure at baseline (Table 2).

Table 2

MOS-HIV Summary and Scale Scores

For almost all summary and scale scores, HRQoL showed an improvement from baseline through to week 24 for both treatment arms. Furthermore, differences between the enfuvirtide + OB group compared with the OB group generally showed a trend toward greater improvement in the enfuvirtide group, and at each time point between 1 and 4 of the differences were statistically significant. For no score at any of the time points did the OB group show a statistically significant improvement over the enfuvirtide + OB group.

At week 4 (Table 3), all scores showed an improvement in HRQoL from baseline with the exception of the social function scale score for the enfuvirtide + OB group (-1.20). The improvement in HRQoL was greater in the enfuvirtide + OB group compared with the OB-alone group for 4 of the 10 scale scores and both summary scores and the difference was statistically significant for the general health scale score (P = 0.02, favoring the enfuvirtide + OB group).

Table 3

At week 8 (Table 3), again almost all scores showed an improvement in HRQoL from baseline, this time with the exception of the general health scale score in the OB-alone group. Scores were higher for the enfuvirtide + OB group compared with the OB-alone group for 9 of the 10 scale scores and both summary scores, and a statistically significant difference was found for the general health scale score (P = 0.01, favoring the enfuvirtide + OB group), as at week 4.

The statistically significant difference between the general health scale score for enfuvirtide + OB compared with OB alone was greater at week 16 (P < 0.01), with the OB group showing a greater decrease from baseline in this scale (-1.68 at week 16 compared with -0.05 at week 8). Eight of 10 scale scores and both summary scores showed a trend toward greater improvement in the enfuvirtide + OB arm than the OB. Similar to week 4, in the enfuvirtide + OB group there was a small decrease in HRQoL from baseline in the social function scale score (-0.10).

Finally, at week 24, for 8 of 10 scale scores and both summary scores in the enfuvirtide + OB group, the trend was toward a greater increase from baseline compared with OB alone and the difference between the treatment arms was statistically significant for 5 of these scales (mental health summary P = 0.02, general health P < 0.01, energy/fatigue P = 0.04, health distress P = 0.01, and quality of life P = 0.04). As at weeks 8 and 16, at week 24 the OB-alone group showed a decrease in general health from baseline (-0.05, -1.68, and -1.36, respectively) as well as in quality of life (-0.27 at week 24). The enfuvirtide + OB group showed a decrease in score for social function, as at week 4 (-1.20 and -0.31, respectively).

At all time points, the general health scale score showed the largest between-group difference, being 3.69, 4.12, 5.82, and 5.29 MOS-HIV points at 4, 8, 16, and 24 weeks, respectively. The health distress scale score exhibited the greatest on-study change in enfuvirtide recipients. There was a clear trend toward improvement in this parameter in the enfuvirtide + OB group compared with the OB group as this was worse in the former group at 4 weeks, better at 8 and 16 weeks, and significantly better at 24 weeks.

DISCUSSION

This study evaluated the impact of therapy containing enfuvirtide + OB and OB therapy alone on HRQoL outcomes, measured using the MOS-HIV, in heavily treatment-experienced HIV-1-infected patients participating in 2 ongoing phase 3 clinical trials. The findings demonstrate that the addition of enfuvirtide to an OB regimen does not adversely affect HRQoL when self-administered for up to 24 weeks.

For most scales, an improvement in HRQoL was seen in both the enfuvirtide + OB group and the OB group at all time points. Although this study was not designed to compare changes in MOS-HIV scores over time, it is interesting to note that most of these HRQoL measures appeared to improve on-study in both treatment groups (as reflected by positive values of the change in least-squares mean from baseline; Table 3). This suggests that optimizing background therapy prior to enrollment into a clinical trial, in itself, tends to improve HRQoL. Therefore, it would be expected that enfuvirtide, in addition to an individualized background regimen, should provide an even greater added benefit to patients when measuring HRQoL outcomes. Indeed, although most of the differences were not shown to be statistically significant, for nearly all scales, the benefit was greater in the enfuvirtide + OB group and for no scale at any time point was there a statistically significant difference showing a trend toward a greater improvement in the OB group.

At week 24, 8 of 10 scale scores and both summary scores showed a greater improvement for the enfuvirtide + OB group compared with the OB group. Four of the 10 scale scores as well as the mental health summary score showed a statistically significant difference in change from baseline compared with OB alone, the highest number of any time point, suggesting that the difference between the 2 arms was most evident at week 24. The general health scale score showed the largest between-group difference of all the evaluated parameters and was statistically significant at all postbaseline time points with differences in favor of enfuvirtide. This scale attempts to capture patients' overall impression of their health by asking patients to rate their health in general as well as how true various statements regarding their health are such as, I am somewhat ill and I am as healthy as anybody I know. As is standard in HRQoL studies with multidomain instruments,^11-14 we did not adjust for multiple comparisons.

Social function appeared to be the parameter least favorably affected by the addition of enfuvirtide. In the enfuvirtide + OB group, this was the only parameter to show a negative change in least-squares mean from baseline (weeks 4, 16, and 24). By contrast, there were no negative changes in this parameter in the OB group. Nevertheless, none of the between-group differences in social function reached statistical significance at any time point.

Certain limitations of this study should be considered. Patients were not blinded to their treatment arm because enfuvirtide is an injectable. Patients were also aware of their clinical progress (CD4⁺ cell counts and viral load) during the study, and this could have influenced their perceptions of HRQoL. However, to address this, the MOS-HIV was administered prior to any clinical procedures or consultations at each visit.

This study did not seek to determine the causes of any HRQoL benefits that might result as a consequence of commencing enfuvirtide. However, it could be speculated that HRQoL benefits of commencing enfuvirtide might be related to virologic and immunologic responses, which showed greater improvement from baseline in the enfuvirtide + OB group compared with OB alone in both TORO trials.^6-9 Prior studies have shown that HRQoL scores are correlated with CD4⁺ cell counts and viral load.^8,11,28 The decrease in social function scale score may be related to the inconvenience of the mode of administration, or possibly related to ISRs, although further studies would be needed to confirm this.

The HRQoL findings in this study are also consistent with the comparable safety profile shown by enfuvirtide + OB compared with OB alone in the TORO trials and earlier studies.^4,5,29-31 Although enfuvirtide therapy is commonly associated with ISRs, low discontinuation rates were observed in this treatment-experienced population.^6-9 In a previous study, improved HRQoL has been shown to be correlated with improved compliance, although the relationship was not specifically measured in this study.¹⁰

Another study evaluating the clinical utility of enfuvirtide used the recently developed Subcutaneous Injection Survey (SIS) instrument.³² In this survey, most patients reported that the preparation and injection of enfuvirtide was easy to perform and had little or no impact on daily functioning and activities of daily living.³³ Unlike the present study, the SIS was specifically designed to assess the impact of subcutaneous injection and did not include a control group. Nevertheless, it suggests that among motivated patients who receive adequate support, patients accepted self-injection of enfuvirtide and this acceptance did not decline over 24 weeks of therapy.

The findings of the present study add to this observation by indicating that self-injected enfuvirtide added to an OB regimen does not adversely affect HRQoL when self-administered for up to 24 weeks. Furthermore, the use of enfuvirtide may even increase some aspects of HRQoL compared with an OB regimen alone. Additional analyses are planned once 48-week data are available.