HIV Articles  
Tesamorelin & IGF-1 Levels
  Reported by Jules Levin
In summary, treatment with a fixed tesamorelin daily regimen of 2 mg had the following effect on serum IGF-1 levels: (see full FDA discussion from their Briefing Document below)
".......suspected link between high IGF-1 levels and the risk of tumorigenesis...." "All analyses indicated a statistically significant elevation in mean IGF-1 at Week 26 (p<0.001)."
"At the start of the Extension Phase (after 26 Weeks of treatment with tesamorelin), mean IGF-1 had increased by 93.94% in the T-T group and by 100.52% in the T-P group. After 13 weeks of the Extension Phase (Week 39 of the trials), the mean IGF-1 change from baseline held steady in the T-T group (+73.15%), whereas patients in the T-P group had experienced a reversal of the IGF-1 increase they experienced in the Main Phase (mean IGF-1 change from Week 0 of -2.40%) This pattern held through Week 52, with a mean IGF-1 increase of 63.07% and a decrease of 9.07% for T-T and T-P groups respectively (p<0.001)."
· It increased the mean serum IGF-1 SD score above the upper limit of normal at 6 months (observation made in two independently conducted studies). The changes occurred as early as 13 weeks of treatment (the earliest timepoint measured in the trial), but given the pharmacodynamic characteristics of the drug they are likely to have occurred earlier. Almost half of the patients treated had IGF-1 SD scores above the upper limit of normal and more than 1/3 had levels greater than +3 SD.
Female patients had a lesser IGF-1 SD elevation, while male patients experienced an even larger increase in mean serum IGF-1.
· Patients who continued tesamorelin for 52 weeks had mean IGF-1 levels in the upper normal range but even in this group of patients as many of 1/3 had SD scores above upper limit of normal and about 1/5 had levels greater than 3 SD; a larger proportion of males had above upper limit elevations when compared to females. An analysis including only extension phase completers suggests that patients who dropped out may have confounded the results and that IGF-1 levels may be expected to be even higher than those recorded at the end of the 52 week trials.
· Since a significant percentage of patients were not fully compliant with the treatment, it is likely that treatment-compliant patients may reach even higher IGF-1 SD scores; this finding is a safety concern given the fact that evidence is accumulating that HIV patients are at higher risk of non-AIDS defining malignances.
· Discontinuation of tesamorelin resulted in a decrease in serum IGF-1 to baseline levels. However, discontinuation of tesamorelin also results in a reaccumulation of VAT.
From: Dragos Roman, M.D. Acting Lead Medical Officer, Division of Metabolism and Endocrinology Products Mary Parks, M.D. Director Division of Metabolism and Endocrinology Products
"It should be emphasized though that, despite the above mentioned rationale, it is not known if VAT reduction with therapies that target the GH axis is associated with any improvements in clinical endpoints such as reduction in number or severity of cardiovascular events or cardiovascular death"
"The information accumulated in the above-listed studies seems to indicate that supraphysiological rhGH doses have an effect on VAT reduction but unfortunately an unacceptable safety profile and, overall, an unfavorable risk-to-benefit analysis, while low doses tend to loose some efficacy but in the process are not devoid of unfavorable effect on glucose metabolism. It should also be noted that regardless of the rhGH doses evaluated, there is no maintenance effect of treatment upon discontinuation. Patients have reaccumulation of VAT to near baseline values upon withdrawal of rhGH therapy."
"Although it has been suggested that natural secretagogues such as GHRH and its analogs may be able to restore a more physiologic pattern of GH secretion because they may preserve both GH secretory pulsatility and IGF-1 feedback inhibition, the observations made in some clinical trials do not clearly support this assumption, as patients treated with these agents do not appear to be free of the adverse events seen with rhGH in general."
"Egrifta, VAT reduction and Cardiovascular risk
In this context, this advisory committee panel is considering the efficacy and safety of Egrifta (tesamorelin acetate), a GHRH analog. This drug is chemically GHRH with a minor modification (an addition of a hexenoyl moiety to the N-terminal part of the sequence) aimed at prolonging its half-life while maintaining the same binding affinity as GHRH in vitro. When injected at a daily dose of 2 mg subcutaneously it demonstrated a statistically significant reduction in VAT at 6 months of 19.6% and 11.7%, respectively, relative to placebo in two independent Phase 3 clinical trials (VAT was measured as the cross-sectional area of a single slice CT scan at the L4-5 level). Egrifta was also associated with other metabolic benefits relative to placebo, such as a reduction in total fat mass of approximately 1.4 kg (almost all of it being due to trunk fat reduction) and an almost identical increase in lean body mass. Other desirable effects, such as triglyceride or cholesterol reduction, improvement in patient related outcomes, and reduction in waist circumference were seen either inconsistently during the trials or were of small magnitude.
Similar to the rhGH trials, the effect of Egrifta on VAT is not sustained with discontinuation of therapy. Patients in the Phase 3 trials who were switched from Egrifta to placebo after 24 weeks demonstrated a reaccumulation of VAT to near baseline levels. This is an important observation in the risk-benefit evaluation as it appears chronic therapy is necessary to maintain reductions in VAT that will also be accompanied by long-term effects of GH and IGF-1 stimulation.
Importantly, the Phase 3 program was not designed to evaluate the effect of Egrifta on CV risk reduction. As stated earlier, one of the rationale for treating HIV-associated lipodystrophy and the metabolic derangements of this condition is to reduce cardiovascular risk. This program has no evidence directly linking the change observed on a single slice CT scan measurement of VAT at L4-5 level or with other endpoints to a reduction of the risk of cardiovascular events such as myocardial infarctions or strokes. Given that the putative link between VAT reduction and cardiovascular benefit has not been validated to date in drug intervention trials (and thus cannot be extrapolated to the Egrifta program) and that no clinical endpoints have been evaluated in the Phase III trials, Egrifta’s treatment effect on secondary endpoints such as triglycerides, cholesterol, patient-reported outcomes, and abdominal circumference have been planned as supportive evidence of efficacy beyond VAT reduction. With only modest and inconsistent results in these secondary endpoints, we are left to consider almost exclusively the benefit of VAT reduction and rely on an unvalidated biomarker for a yet-to-be demonstrated cardiovascular benefit.
In the final analysis, even if the degree of VAT reduction and other favorable metabolic changes promoted by Egrifta were to be considered valid surrogates of clinical benefit, one needs to make a determination as to what is the benefit-to-risk ratio for Egrifta in the face of 1) a statistically significant increase in the number of patients with diabetes in the tesamorelin group relative to placebo (a finding that may affect unfavorably the potential cardiovascular benefit of this therapy), and 2) an increase in the serum IGF-1 values above the upper range of normal in a considerable number of Egrifta-treated patients, especially since this treatment is anticipated to be given long-term and acknowledging the fact that HIV patients are at risk of non-AIDS defining malignancies."
Main Phase

IGF-1 measurements were performed at baseline, Week 13 and Week 26. Mean baseline IGF-1 SD scores were within the low normal range: -0.31 for the tesamorelin and -0.21 for the placebo group, respectively. Small differences in mean baseline IGF-1 levels existed between Study 10 (SD score close to 0.00) and Study 11 (SD score of about -0.4). The vast majority of patients had IGF-1 levels below the upper limit of normal (i.e. < 2 SD), with only 6% of patients displaying IGF-1 SD scores above the normal range (i.e. > 2 SD) at baseline.
At Week 26, the mean IGF-1 SD score increased above the upper limit of normal (2.39) in the tesamorelin group while for the placebo groups it remained in the normal range and below the study population mean (-0.45). Changes at Week 13 were consistent with those seen at Week 26 (Table 30). The percentage of patients with IGF-1 SDS values above the upper limit of normal increased from 6.2 % at baseline to 47.4 % in the tesamorelin group and remained virtually unchanged in the placebo group (6.1 % at baseline and 5% at Week 26). Moreover, the percentage of patients with SD scores above 3 standard deviations increased from 1.5% at baseline to 35.6% in the tesamorelin group with no real change in the placebo group (3.8% at baseline and 2.5% at week 26).
Also of interest is the potential effect of non-compliance on the IGF-1 data. For instance, in Study 10 non-compliance (defined in the protocol as actual administration of <80% of scheduled doses) was found in 26.2 % of patients, while in Study 11 it was 39.5%. This observation indicates that in compliant patients IGF-1 levels may be even higher.
A breakdown by gender of the IGF-1 data is provided in Table 31. A significantly greater number of males compared to females enrolled in the study (457 vs. 77 in the tesamorelin group; 219 vs. 42 in the placebo group). At baseline the mean IGF-1 SD score was lower in females than males (approx. -0.7 females vs. -0.2 males). Regardless of gender, most patients had IGF-1 SD scores below the upper limit of normal with only a few exceeding it. At Week 13, the mean SD score for males receiving tesamorelin increased to 2.70 compared with only 1.13 for females. At Week 26, the means were similarly higher in males (2.62) versus females (0.94). Furthermore, a higher proportion of males had SD scores above 2 or 3 standard deviations when compared to females. Specifically, 52.4 % and 51.0% of males in the tesamorelin group had an SDS score >2 at Weeks 13 and 26, respectively, compared with only 27.9% and 24.1% of females; 37.0% and 38.2% of males had SDS scores >3 at Weeks 13 and 26, respectively, compared to only 14.8% and 18.5% of females. This, coupled with higher changes from baseline seen in males, indicates a clear gender-specific IGF-1 response with tesamorelin.
These data also indicate that the peak IGF-1 level is reached in both genders by Week 13 (the earliest post-baseline assessment). Given the known pharmacodynamic profile of IGF-1 following the administration of exogenous rhGH, it is very likely that such levels may be reached well before Week 13 (even within days) suggesting that patients are exposed to the levels of IGF-1 observed at Weeks 13 and 26 throughout most of the six-month trial.
Extension Phase
During the extension phase the mean IGF-1 SDS decreased in the T-T group from 2.66 at Week 27 to 2.13 at Week 39 and 1.70 at Week 52. This change happened in the context of a concomitant reduction in the number of patients who contributed measurements to this analysis from 236 at Week 27 to 190 at Week 52.
The percentage of patients in the T-T group with IGF-1 measurements above 2 SD decreased from 50% at Week 27 to 33.7% at Week 52, as did that of patients with IGF-1 SD score >3, from 39.8% to 22.6 %, respectively. However, despite the reduction in mean IGF-1 levels and the decrease in the percentage of patients with above normal IGF1 levels at Week 52, as many as 1/3 patients had IGF-1 levels > 2 SD and more than 1/5 had levels > 3 SD after one year of treatment. In contrast, patients in the T-P group who completed 52 Weeks of treatment had a reduction in mean IGF-1 SD score from 2.27 at Week 27 to values close to those recorded at the trial initiation (-0.58). All the findings described above are summarized in Table 32.
Finally, the patients in the P-T group (not included in Table 32), reproduced to a large extent the findings of the tesamorelin group during the Main Phase of the trials. Although the mean IGF-1 SDS values did not go above the upper limit of normal (-0.42 at baseline and 1.69 at end of the 6 months of treatment), the percentage of patients with values > 2 SD increased from 5.2% at baseline to 41.1 % at end-of-trial, as did the percentage of patients with values > 3SD which increased from 2.6% to 29.1% for the same duration of treatment.
An analysis of IGF-1 levels by gender is presented in Table 33. Patients of both genders in the T-T group experienced a lowering of mean IGF-1 SDS scores over the course of the Extension Phase (from 2.87 at Week 27 to 2.26 at Week 39 and 1.87 at Week 52 in males; from 0.89 to 1.01 to 0.38 for the same timepoints in females). Similar trends were noted for the proportion of patients in the T-T group with SDS scores >2 or >3. However, a sizeable proportion of male patients maintained above-normal IGF-1 SD scores at Week 52: 36.3 % had SDS > 2 and 25.5% > 3 SDS. In contrast only 13.6% of females had a SD score > 2 SD and none > 3 SD for the same timepoint.
Analyses of IGF-1 SD scores restricted to Extension Phase completers Since the reduction in mean IGF-1 levels at Week 52 could have been confounded by the fact that some patients discontinued the trial for various reasons (and some of them may have had excessively high IGF-1 levels), the FDA statistical reviewer has conducted several analyses that exclude dropouts and focus only on the patients who had received treatment and had trial participation through Week 52. These patients are presented in Figure 9, which presents the mean IGF-1 SD scores in Studies 10 and 11/12 side-by-side. The element of immediate interest in the graph is the blue line that describes the mean IGF-1 SDS for patients who received tesamorelin through Week 52 and completed the trial. The trends observed are very similar to those described previously in that the mean IGF-1 SDS increased above the upper limit of normal, peaked at Month 6, and decreased subsequently. There were, however, some quantitative differences. In Study 10, the mean (SD) IGF-1 SDS at Week 52 was 1.6 (2.2) with a range between -3.5 and 10.5; in Study 12 it was higher at 2.3 (2.8) with a similar range (-3.1 to 11.6), suggesting an average value for the two studies combined close to 2 SD at Week 52 and higher than observed in Table 32. It should be mentioned that these analyses do not account for the potential confounding effect of non-compliance. Non-compliance in Study 10-extension was 31.3% and it was even higher in Study 12 (50%). This suggests that, had patients been fully compliant, they could have conceivably had even higher mean IGF-1 SD scores (per protocol, a patient was deemed non-compliant if he received <80% of the planned study drug doses).
When the IGF-1 SDS analyses conducted on the extension-phase completers were broken down by gender, the pattern observed was similar to that previously described. For female patients the mean (SD) scores at Week 52 were in the upper range of normal: 0.7 (1.3) with a range between -1.4 and 3 in Study 10; and 1.3 (1.1) with range of 0.2 to 2.9 in Study 11. For males the mean values at Week 52 were higher when compared to females and they also had a wider range. The mean (SD) was 1.7 (2.3) with a range of 3.5 to 10.5 in Study 10, and 2.5 (2.9) with a range of -3.1 to 11.6 in Study 11. This information is displayed graphically in Figure 10. As before, the blue line represents the group of interest (the T-T group). Females are represented in the upper two panels and males in the lower two panels. .
  iconpaperstack view older Articles   Back to Top