mTOR inhibitors in the pharmacologic management of tuberous sclerosis complex and their potential role in other rare neurodevelopmental disorders

Experience with sirolimus in treating SEGA was evaluated in case reports and as a secondary end point in a phase two trial with a small number of patients. In these cases, sirolimus demonstrated an observable regression of SEGA lesions [30‐32].

Everolimus has been studied more extensively in treating SEGA through long-term phase 2 and 3 studies [33‐36]. In a 6-month open-label phase 2 study consisting of 28 patients, everolimus demonstrated a significant reduction in tumor volume compared with baseline, with approximately 75% of patients experiencing a ≥ 30% reduction in SEGA volume and 32% experiencing a ≥ 50% reduction [33]; these reductions were sustained during the extension phase of the trial (median 5.65 years of treatment) [37]. In a randomized, double-blind, placebo-controlled, phase 3 study of 117 patients with SEGA associated with TSC, treatment with everolimus (median 9.6 months) was associated with a significantly higher SEGA response (≥ 50% reduction of SEGA volume) rate compared with placebo (35% vs. 0%; p < .0001) [35]. An analysis of 111 patients who received at least one dose of everolimus (in either the double-blind or a subsequent open-label phase) revealed that SEGA response increased to 57.7% over a median duration of 47.1 months (3.9 years), and the median reduction in SEGA volume was maintained, and even slightly increased, over the duration of the study [38]. Taken together, phase 2 and 3 clinical data on everolimus supported its use in the setting of TSC-associated SEGA, with the phase 2 results leading to the US Food and Drug Administration (FDA) approval of everolimus for the treatment of SEGA in pediatric and adult patients with TSC [39].

Everolimus was evaluated for the management of renal angiomyolipoma in the large phase 3 EXIST-2 trial and in a subset of the patients from the EXIST-1 trial who had SEGA and renal angiomyolipoma [40, 41]. In EXIST-2, the angiomyolipoma response rate (≥ 50% reduction in volume in absence of other factors) after approximately 8 months of treatment was 42% for patients taking everolimus compared with 0% in patients receiving placebo (p < .0001) [40], which increased to 54% in patients treated with everolimus for a median of 29 months [42], and 58% at the completion of the open-label extension phase (median exposure, 46.9 months) [43]. Based on the results from the core phase of EXIST-2, everolimus was approved by the FDA for the treatment of adult patients with renal angiomyolipoma and TSC [39]. Similar to the findings of EXIST-2, a subset of patients with SEGA and angiomyolipoma in EXIST-1 (largely pediatric population) reported angiomyolipoma response rates of 53.3% for everolimus and 0% for placebo after a median of 9.6 and 8.3 months of treatment, respectively; 80% of patients achieved a ≥ 50% reduction in renal angiomyolipoma volume after 48 weeks (11 months) of treatment [41].

Sirolimus has not been approved for the management of renal angiomyolipoma, but has been evaluated in several small open-label phase 2 clinical studies [32, 44‐46]. Bissler et al. found that sirolimus reduced the size of angiomyolipoma lesions and improved lung function over 12 months of treatment [44]. However, 12 months after sirolimus was discontinued, lesion size and several lung function parameters approached baseline levels, suggesting that therapy with mTOR inhibition might necessitate long-term or indefinite use [44]. Davies et al. conducted a longer study and found that 50% of patients reported a positive angiomyolipoma response (disappearance of lesions or ≥ 30% reduction in sum of longest diameter of target lesions) over a 2-year period [45]. One phase 2, multicenter trial of sirolimus in adults with TSC evaluated the effects of sirolimus on multiple lesion types. Over a period of 1 year of treatment, they observed reductions in renal angiomyolipoma size, SEGA size, and liver angiomyolipoma size, with subjective improvement in skin lesions and reduction of vascular endothelial growth factor (VEGF) D [32].

Sirolimus and everolimus have both been evaluated for the management of LAM in a number of studies consisting primarily of patients with sporadic LAM, although small numbers of patients with TSC-associated LAM were also included [47‐50]. In the multicenter placebo-controlled MILES study, 89 patients with LAM (8 with a codiagnosis of TSC) receiving treatment with sirolimus (n = 46) over 12 months exhibited improvements in forced vital capacity (FVC) and quality of life, as well as stabilization of forced expiratory volume in 1 second (FEV₁) [47]. These findings led to the FDA approval of sirolimus for the treatment of LAM [51]. Two retrospective studies also evaluated sirolimus in treating LAM [48, 49], reporting improved or stabilized lung function even at serum trough levels < 5 ng/mL [48], along with sustained effects over a treatment period of approximately 3.5 years [49].

Recently, everolimus was evaluated in a prospective study that included 24 patients (5 with TSC-LAM) and showed improvements in FEV₁, stabilization of FVC, and reductions in VEGF-D and collagen IV; however, optimal dosing of everolimus for this indication needs further investigation [50]. As a result, everolimus has yet to receive approval for use in the LAM setting.

Although no mTOR inhibitors are currently indicated specifically for the treatment of seizures associated with TSC, recent clinical evidence has shown promise for this use in this setting. The results of several small reports suggest that sirolimus may be effective for the treatment of TSC-associated seizures [52‐54]. Sirolimus therapy given over 10 months in a 10-year-old girl reduced daily seizure activity from 5–10 times/day to 1–5 times/day and resulted in the cessation of seizure clusters [52]. A case series of seven children with TSC found that all patients experienced seizure control after 12 months of treatment with sirolimus [53]. A second case series of seven children with TSC and refractory seizures reported that most patients had 50% to 90% reductions in numbers of seizures [54]. In a recent, small, randomized trial of 23 children (ages 3 months to 12 years) with TSC, treatment with sirolimus decreased overall seizure frequency 41% over standard of care, but this change failed to reach statistical significance (p = .11) [55].

Use of everolimus in TSC-associated refractory seizures has also been evaluated [33, 56]. A prospective, phase 1/2 trial directly evaluating everolimus in managing refractory seizures associated with TSC showed a reduction in seizure frequency of ≥ 50% in 12 of 20 patients after 12 weeks of treatment [56]. In a phase 2 study, everolimus therapy was associated with a clinically relevant reduction in the overall frequency of clinical and subclinical seizures (median change, −1 seizure; p = .02) in patients with SEGA. Of the 16 patients for whom electroencephalographic data were available, seizure frequency decreased in nine patients after 6 months; five additional patients did not experience an event [33]. Results from the first phase 3 study to evaluate an mTOR inhibitor (everolimus) for refractory seizures associated with TSC were recently reported (ClinicalTrials.gov NCT01713946) [57]. This prospective, randomized, double-blind, multicenter study compared everolimus at two different trough levels (low exposure, 3–7 ng/mL; high exposure, 9–15 ng/mL) with placebo in reducing seizures (N = 366) when added to an existing antiepileptic drug regimen. After 18 weeks of treatment, the median percentage reduction in seizure frequency was significantly higher with everolimus (29.3% for everolimus low exposure and 39.6% for everolimus high exposure compared with 14.9% with placebo [p = .0028 and p < .0001, respectively]), and the proportion of responders (≥ 50% reduction in seizure frequency) was significantly greater with everolimus (28.2% for everolimus low exposure and 40% for everolimus high exposure compared with 15.1% with placebo [p = .0077 and p < .0001, respectively]) [57]. These preliminary findings indicate that adjunctive treatment with everolimus may be an effective option in reducing refractory seizures in patients with TSC.

mTOR inhibitors may also be a rational candidate for the management of neurodevelopmental/neuropsychiatric disabilities associated with TSC, including intellectual disability and autism. Indeed, a recent preclinical study of adult rats with TSC2 mutations and developmental status epilepticus, and a case study of a patient with TSC both reported improvements in social deficit behaviors, including autism-related behaviors, following mTOR inhibitor therapy with everolimus [58, 59]. However, mTOR inhibitors have not been adequately evaluated or approved for the treatment of neurodevelopmental disabilities in TSC, especially in young infants. It is also essential that we establish the safety and overall impact of mTOR inhibitors in the pediatric population before larger, definitive clinical trials can be pursued. In the future, we await further information on effects of mTOR inhibitors on TSC-associated neuropsychiatric disorders, including secondary analyses from EXIST-3, and results from several phase 2 trials (NCT01289912, NCT01954693).

In Leigh syndrome, genetic defects result in disruption of mitochondrial function, which contributes to numerous health problems. Patients can have symptoms such as respiratory abnormalities, ocular and other cranial nerve palsies, involuntary movements, motor delays, intellectual disabilities, and seizures [60]. Although the time of onset can vary, it typically occurs in the first year of life. Leigh syndrome is characterized by diffuse multifocal spongiform degeneration in various parts of the brain, and many patients die within a few years after symptom onset [60].

In a preclinical study with Ndufs4 knockout mice (the protein product of the Ndufs4 gene is involved in the assembly, stability, and activity of complex I of the mitochondrial electron transport chain), rapamycin administration increased survivability and health [61]. The mechanism behind this is not entirely understood; however, it is believed that reduction of mTOR activity may shift cell metabolism toward amino acid catabolism and away from glycolysis and, thus, reduce the buildup of glycolytic intermediates that are associated with Leigh syndrome [61].

Additional research has suggested mTOR inhibition may aid in Leigh syndrome through preservation of adenosine triphosphate (ATP). Mitochondria provide energy to the cell through ATP, which has been found to be decreased in mitochondrial disorders; this leads to the degeneration of neurons, as in Leigh syndrome [62]. In an in vitro study, rapamycin was introduced to neuronal cells with mitochondrial defects, resulting in a significant rise in ATP level while protein production slowed [62]. It is theorized that the decrease in the energy-consuming process of protein synthesis with mTOR inhibition allows for more ATP to be spared [62].

Although investigation into the use of mTOR inhibitors in Leigh syndrome is at a very early stage, preclinical results are promising because there are currently no effective therapies for this disease.

Down syndrome is a genetic disorder associated with intellectual disability caused, in most cases, by trisomy of human chromosome 21 [63]. Down syndrome is characterized by abnormalities in dendritic morphology and synaptic plasticity, and mTOR is believed to be involved in the growth and branching of dendrites in the hippocampus [64]. The mTOR activity of dendrites in the hippocampus has been shown to be increased in a mouse model of Down syndrome [64]. This increase was subsequently reversed after administration of rapamycin. Studies are underway to investigate whether rapamycin can reverse learning deficits associated with Down syndrome.

Hyperactivation of the PI3K/Akt/mTOR pathway was also observed in autopsy samples of patients with Down syndrome compared with controls [65]. A causative factor of Down syndrome is hypothesized to be the triplication of amyloid-beta protein gene, resulting in excess protein proliferation. In combination with decreased autophagy as a result of increased mTOR activation, this may result in the accumulation of amyloid beta peptide in the brain and contribute to the neurodegenerative process and to eventual Alzheimer’s-like dementia in these patients [65]. Oxidative stress is also believed to have a role in neurodegenerative diseases such as Down syndrome. A mouse model of Down syndrome demonstrated that protein oxidation was increased possibly due to the decreased protective effect of autophagy as a result of mTOR pathway hyperactivation [66]. Signs of protein oxidation in cells were reduced when rapamycin was introduced [66].

Neurofibromatosis is an autosomal dominant genetic disorder that is further classified into subtypes 1 and 2. Neurofibromatosis types 1 and 2 are caused by inactivating mutations in NF1 and NF2 genes, respectively [67]. The loss of NF1 encodes for the protein neurofibromin and results in development of neurofibromas on or around peripheral nerves, along with pigmented tumors of the skin and iris [67]. Plexiform neurofibroma occur in up to one-third of individuals with neurofibromatosis type 1 and can cause disfigurement, compression of other bodily structures, neurologic dysfunction, and pain [68]. Evidence suggests that neurofibromin is involved in negatively regulating the mTOR pathway. A phase 2 study involving patients with progressive plexiform neurofibromas treated with sirolimus showed a modestly increased time to progression [68]. However, a similar phase 2 study that evaluated sirolimus with nonprogressive plexiform neurofibromas showed that sirolimus did not cause any tumor shrinkage [69]. A case series involving patients with symptomatic plexiform neurofibromas showed that, although sirolimus did not shrink tumor volume, pain was alleviated [70]. The lack of an antitumor response with sirolimus in neurofibromatosis type 1 may be caused by alternative compensatory mechanisms (e.g., feedback activation of Akt activity) following mTOR inhibition [71].

Neurofibromatosis type 2, the rarer of the two subtypes, involves the loss of the NF2 gene, which encodes for the regulator protein merlin. The loss of merlin leads to development of benign tumors called schwannomas, which can grow along auditory nerves, leading to hearing loss, and can compress nerves, leading to increased intracranial pressure, nerve dysfunction, and pain [67]. Similar to neurofibromin, merlin has been found to be a negative regulator of the mTOR pathway [72]. Rapamycin arrested schwannoma tumor growth in mice and in an in vitro model [72]. However, in a phase 2 study evaluating everolimus in the treatment of progressive vestibular schwannomas, none of the patients experienced a response (≥ 15% reduction in tumor volume) [73]. The activation of negative feedback loops following mTOR inhibition may also explain the limited efficacy of everolimus in vestibular schwannomas [73].