Abstract
Treatment of locally advanced basal cell carcinomas (laBCCs) with large, aggressive, destructive, and disfiguring tumors, or metastatic disease is challenging. Dysregulation of the Hedgehog (Hh) signaling pathway has been identified in the vast majority of basal cell carcinomas (BCCs). There are two United States Food and Drug Administration (US FDA)-approved Hh pathway inhibitors (HPIs) that exhibit antitumor activity in advanced BCC with an acceptable safety profile. Common adverse effects include muscle spasms, dysgeusia, alopecia, fatigue, nausea and weight loss.
Introduction
More than 2.8 million new cases of basal cell carcinoma (BCC), the most common human cancer, are diagnosed each year in the United States and contribute to over 3000 deaths [Madan et al. 2010; Mohan and Chang, 2014]. Most BCCs are effectively treated with surgical excision, Mohs surgery, destruction or topical therapy. A study of facial BCCs found 5-year recurrence rates after excision or Mohs surgery were 2.5–4.1% for primary tumors, and 2.4–12.1% for recurrent tumors [Mosterd et al. 2008]. When BCC is not promptly treated, extensive local tissue invasion and destruction in addition to rare occurrence of metastasis can lead to significant morbidity and mortality. Surgery and radiation therapy to treat advanced cases may cause unacceptable disfigurement or morbidity.
The pathogenesis of BCC has been elucidated including identification of aberrant Hedgehog (Hh) pathway signaling and the associated mutations of Patched-1 (PTCH1) and Smoothened (SMO) genes [Gailani et al. 1996; Hahn et al. 1996; Johnson et al. 1996; Aszterbaum et al. 1998; Xie et al. 1998; Epstein, 2008]. Aberrant Hh pathway signaling has been detected in patients with nevoid BCC syndrome as well as in over 95% of patients with sporadic BCCs [Gailani et al. 1996; Xie et al. 1998; Epstein, 2008]. The discovery of receptor-targeted molecules in the Hh pathway has led to the approval of two Hh pathway inhibitors (HPIs) vismodegib and sonidegib by the United States Food and Drug Administration (US FDA) (Figure 1).
When Hh signaling ligand is present and binds to the 12 transmembrane protein PTCH1, the 7 membrane spanning receptor SMO is no longer inhibited via PTCH1-mediated inhibition and allows transcription factor Gli1 to enter the cell nucleus, stimulating cell division and tumorigenesis. Both vismodegib and sonidegib bind SMO and inhibit the Hh pathway reactivation.
Vismodegib (Erivedge Capsule, Genentech, Inc., USA) was approved by the US FDA in January 2012 for the treatment of adults with metastatic basal cell carcinoma (mBCC) or with locally advanced basal cell carcinoma (laBCC) that has recurred following surgery or who are not candidates for surgery or radiation. The US FDA approval of vismodegib was based on the demonstration of objective response rate (ORR) from the ERIVANCE clinical trial, a multicenter, single-arm, two-cohort, clinical trial, evaluating the treatment of 150 mg of vismodegib daily in patients with laBCC or mBCC [Sekulic et al. 2012].
Sonidegib (Odomzo Capsules, Novartis Pharmaceuticals Corp., USA) was approved by the US FDA in July 2015 for the treatment of patients, 18 years or older, with laBCC that has recurred following surgery or radiation therapy, or those who are not candidates for surgery or radiation therapy. The US FDA approval of sonidegib was based on the demonstration of durable ORR from the BOLT clinical trial, a phase II, multicenter, randomized and double-blinded clinical trial, evaluating the treatment with two different doses of sonidegib in patients with laBCC or mBCC [Migden et al. 2015].
This article will provide an overview of clinical use of HPIs in advanced BCC.
Drug development
The Hh pathway was discovered during the investigation of Drosophila melanogaster embryogenesis [Nusslein-Volhard and Wieschaus, 1980]. The activation of the Hh pathway during embryogenesis ensures proper development of the embryonic cells; however, the Hh pathway remains quiescent in adult tissues. In the absence of the Hh ligand signal, the cell-surface transmembrane protein PTCH-1 keeps the activity of the 7-membrane-spanning receptor, SMO, suppressed [Stone et al. 1996; Xie et al. 1998]. However, when the Hh signaling ligand is present and bound to PTCH-1, SMO is no longer inhibited and allows the transcription factor Gli to enter the cell nucleus, promoting cell division and tumorigenesis [Aszterbaum et al. 1998].
The first SMO antagonist reported in literature is a naturally occurring alkaloid called cyclopamine found in the corn lily [Cooper et al. 1998]. Cyclopamine was identified to be the culprit responsible for the midline single-eyed appearance of many lambs born by corn lily-consuming ewes in the 1950s. Cyclopamine was found to bind SMO and inhibits the activation of downstream Hh target genes that are essential for proper embryonic development [Incardona et al. 1998]. In preclinical mouse models, cyclopamine has been shown to induce the remission of medulloblastoma via Hh signaling inhibition [Sanchez et al. 2005]. Despite its ability to interfere with the Hh pathway signaling, cyclopamine’s suboptimal aqueous solubility and chemical stability limit its therapeutic usefulness in clinical research [Winkler et al. 2009].
GDC-0449 (vismodegib) is a first-in-class, small-molecule inhibitor of SMO that selectively inhibits the Hh signaling pathway by binding to SMO, therefore inhibiting the aberrant activation of the Hh pathway. GDC-0449 has greater potency and more favorable pharmaceutical properties than cyclopamine. In 2009, a phase I study of GDC-0449 (vismodegib) showed a 58% response rate among 33 patients with laBCC and mBCC [Von Hoff et al. 2009]. The median duration of response was 12.8 months [Von Hoff et al. 2009]. The recommended dose of vismodegib was 150 mg/day, based on the achievement of maximal plasma concentration and pharmacodynamic response [Lorusso et al. 2011a].
LDE225 (sonidegib), discovered in 2010, is a potent selective SMO antagonist with high tissue penetration and the ability to cross the blood–brain barrier with good oral bioavailability [Pan et al. 2010]. In a phase I study in 103 patients with advanced BCC and other solid tumors, the maximum tolerated dose was determined to be 800 mg daily and 250 mg twice daily after testing various of dosages [Rodon et al. 2014]. No clinical advantage was observed with twice-daily dosing over once daily dosing; therefore, the once daily dosing regimen is currently recommended [Rodon et al. 2014].
LDE225 0.75% topical cream was investigated in 27 BCCs treated twice daily for 4 weeks [Skvara et al. 2011]. In the treatment group (13 of 27); 3 BCCs had a complete response, 9 BCCs had a partial response, and one BCC had no clinical response. A mean tumor volume reduction was 56%. In the control group (14 of 27), 13 BCCs showed no clinical response, and 1 BCC had a partial response. A phase II study with a SLIGHTLY larger sample size showed insufficient efficacy with investigational topical formulation and treatment conditions.
Metabolism and drug interactions
The major metabolic pathways of vismodegib include oxidation, glucuronidation and pyridine ring cleavage [Genentech, Inc., 2012]. Vismodegib elimination involves multiple pathways predominated by the hepatic route as 82% of the administered dose was recovered in the feces and 4.4% was recovered in the urine [Genentech, Inc., 2012]. No change in vismodegib level was observed in patients concomitantly treated with CYP3A4 inhibitors and CYP3A4 inducers. In contrast, sonidegib is mainly metabolized by the hepatic enzyme CYP3A, therefore, it should not be used in patients taking CYP3A inhibitors or CYP3A inducers due to potential drug interaction [Novartis Pharmaceuticals Corp., 2015].
Warning and precaution
HPIs are contraindicated in women who are either pregnant or breastfeeding as it may cause death or severe birth defects in developing fetus because Hh signaling plays a pivotal role in the embryonic developmental pathway. Reproductive toxicology studies in rats demonstrated that vismodegib exposure during organogenesis results in embryo-fetal death at higher exposures and severe birth defects at exposures within the range achieved with the recommended human dose [Genentech, Inc., 2012]. Although no human data on the effects of HPIs on fetal development have been reported, pregnancy status should be verified prior to the initiation of HPIs. Highly effective contraception should be used during treatment and for up to 7 months after the last dose of vismodegib, and for at least 20 months after the last dose of sonidegib. Male patients should use condoms during treatment with HPIs, and for at least 2 months after the last dose of vismodegib, or 8 months after the last dose of sonidegib. Healthcare providers should report to pharmaceutical company any cases of exposure during pregnancy (either direct exposure in female patients or through seminal fluid from male patients), and should encourage pregnant women to participate in the Erivedge or Odomzo pregnancy pharmacovigilance program to collect information on pregnancy outcomes.
Currently there are no data available regarding the presence of HPIs in human milk, however, breastfeeding is not advisable during treatment and for at least 20 months after the last dose of sonidegib. Because both the ERIVANCE BCC and BOLT trials were conducted in the adult population, the safety and effectiveness of HPIs have not been established in pediatric patients. Patients should not donate blood or blood products while taking HPIs and for at least 7 months after the last dose of vismodegib or 20 months after the last dose of sonidegib. Serum creatinine kinase (CK) levels and renal function tests should be obtained prior to initiating and periodically during sonidegib treatment [Novartis Pharmaceuticals Corp., 2015].
Administration and formulations
The recommended dose for vismodegib is 150 mg orally daily, and it may be taken with or without food. In the pharmacokinetic dose-scheduling study of vismodegib, it showed that any reduction in vismodegib dose or frequency could result in suboptimal unbound drug concentration and therefore compromise the maximum clinical benefit. Dose reduction is not indicated for vismodegib [Lorusso et al. 2011b].
The sonidegib dose is 200 mg orally daily taken on an empty stomach, at least 1 hour before or 2 hours after a meal.
Clinical uses of Hh pathway inhibitors
Vismodegib
In the ERIVANCE clinical trial, a multicenter, international, two-cohort, nonrandomized study, a total of 104 patients with mBCC and laBCC who had inoperable disease or for whom surgery was inappropriate were enrolled and received 150 mg of oral vismodegib daily. The primary end point of the ERIVANCE trial was the independently assessed ORR. For laBCC, the objective response was defined as a decrease of 30% or more in the externally visible (including residual scarring) or radiographic dimension, or complete resolution of ulceration if present at baseline. Progressive disease was defined as an increase of 20% or more in the externally visible or radiographic dimension, new ulceration, or a new lesion. For patients with multiple target lesions, the sum of the longest diameters was used to determine response and progression. The Response Evaluation Criteria in Solid Tumors (RECIST) was used for mBCC [Therasse et al. 2000]. Patients received vismodegib until disease progression, unacceptable toxic effects, or discontinuation of the study. The mean duration of treatment was approximately 10 months.
In 63 patients with laBCC, the ORR (including both complete and partial response) was 43%, with complete responses in 13 (21%) patients (defined as the absence of residual BCC on assessment of a biopsy specimen). The ORR of 33 patients with mBCC was 30%, and all responses were partial. The median duration of response was 7.6 months in both groups [Sekulic et al. 2012].
Adverse events occurring in more than 20% of patients were muscle spasms (68%), alopecia (63%), dysgeusia (taste disturbance, 51%), weight loss (46%), fatigue (36%), nausea (29%), decrease in appetite (23%), and diarrhea (22%). Adverse events of grade 3 (severe or medically significant) or grade 4 (life threatening) included muscle spasms, weight loss, fatigue, and loss of appetite [Sekulic et al. 2012]. Of the 104 patients enrolled, 13 (12%) had an adverse event leading to discontinuation of the study drug with muscle spasm being the most commonly reported event that contributed to study drug discontinuation, which occurred in 2 individuals. Serious adverse events of any grade were reported in 26 patients (25%). Grade 5 (fatal) adverse events reported in 7 patients (6 with laBCC and 1 with mBCC) were due to hypovolemic shock, myocardial infarction, meningeal disease and ischemic stroke that were responsible for 1 death each and 3 patients died from unknown causes. These 7 patients had clinically significant risk factors or coexisting conditions at baseline. No association with the study drug was identified.
An additional 12-month follow-up in the phase II, multinational, multicenter, nonrandomized, two-cohort study, confirms the efficacy and safety of vismodegib in the management of advanced BCC [Sekulic et al. 2015]. Median follow-up time, including the primary analysis and the additional 12-month follow up, was 21.7 months in the laBCC group and 22.4 months in the mBCC group. The ORR increased from 42.9% to 47.6% in patients with the laBCC, and from 30.3% to 33.3% in patients with mBCC. All responders (33.3%) in the mBCC group had partial response, and 22.2% and 25.4% of responders in the laBCC group had complete and partial response respectively. Median duration of response in patients with laBCC increased from 7.6 to 9.5 months (Table 1).
Response rate comparison in sonidegib and vismodegib.
laBCC, locally advanced basal cell carcinoma; mBCC, metastatic basal cell carcinoma.
No new safety signals emerged with extended treatment duration. Muscle spasms, dysgeusia and alopecia remained the most common adverse events in both the primary and the 12-month update analysis. Amenorrhea was reported in 2 patients with laBCC. As of the 12-month update, of the 104 patients enrolled, 75 patients discontinued treatment mainly due to patient decision, disease progression and adverse events. The most common adverse events leading to treatment discontinuation included muscle spasms, weight loss and dysgeusia.
Almost half of patients had grade 1 (mild) to grade 2 (moderate) adverse events, and 51.9% of patients had adverse events grade 3 or higher. Serious adverse events (31.7% of patients) increased modestly compared with those at primary analysis (25% of patients). No death occurred as the result of adverse events.
Sonidegib
In the BOLT trial, a multicenter, randomized, double-blind, phase II study, patients with laBCC not amenable to curative surgery or radiation, or mBCC were randomized in a 1:2 ratio to receive sonidegib 200 mg daily (lowest efficacious dose) or 800 mg daily (maximum tolerated once daily dosing). RECIST [Therasse et al. 2000] and modified RECIST was used to assess mBCC and laBCC, respectively [Migden et al. 2015; Eisenhauer et al. 2009]. The primary endpoint was the proportion of patients who achieved an objective response (complete or partial response) with data collected up to 6 months after randomization of the last patient. In the setting of symptoms indicating potential toxicity, dose interruptions or reductions were permitted. Of the 230 patients enrolled, 194 with laBCC and 36 with mBCC, 79 were in the 200 mg sonidegib group, and 151 were in the 800 mg sonidegib group. The median durations of treatment were 8.9 months in the 200 mg dose group and 6.5 months in the 800 mg dose group dose. Median follow up was 13.9 months.
In the primary efficacy analysis population, 20 (36%) of 55 patients receiving 200 mg sonidegib and 39 (34%) of 116 receiving 800 mg sonidegib achieved an objective response per central review. In the 200 mg sonidegib group, 18 (43%) patients who achieved an objective response were noted among the 42 with laBCC and two (15%) among the 13 with mBCC. In the 800 mg sonidegib group, 35 (38%) patients who achieved an objective response were noted among the 93 with laBCC and 4 (17%) among the 23 with mBCC. A reduction in Gli1 expression was seen in patients with disease control (those with either complete response, partial response or stable disease) (Table 1).
The most common adverse events included muscle spasms, alopecia, dysgeusia (taste disturbance), nausea, elevated blood CK and fatigue [Migden et al. 2015]. Muscle spasm was the most common reported adverse event in 49% and 67% of patients in the 200 mg and 800 mg group respectively, and the most common adverse event resulting in treatment discontinuation. Alopecia was reported in 43% of patients in the 200 mg group and 55% of patients in 800 mg group. Taste disturbance was reported in 38% and 59% of patients in the 200 mg and 800 mg group respectively. Adverse events generally occurred less frequently in the 200 mg group than in the 800 mg group.
The most common grade 3–4 adverse events (severe, medical significant or life threatening) were elevated serum CK (6% in the 200 mg group versus 13% in the 800 mg group) and elevated lipase (5% of patients in both groups). Increasing sonidegib dose and exposure were associated with increased CK level. Serious adverse events were reported in 14% of patients in the 200 mg sonidegib group and 30% in the 800 mg sonidegib group. Rhabdomyolysis and elevated serum CK were the most frequently reported serious adverse events which both occurred in 1% of patients in the 200 mg group and 3% of patients in 800 mg group. However, these reported rhabdomyolysis cases did not meet the criteria defined by the Independent Safety Review Committee and Adjudication Committee as CK concentrations more than ten times higher than baseline plus a 1.5-fold increase in creatinine concentration in serum from baseline [Migden et al. 2015]. Secondary malignancies were noted in 6% (200 mg) and 7% (800 mg) of patients. Overall, 4 patients in the 800 mg group died while taking study treatment, but none of these deaths was deemed to be related to sonidegib.
Although both groups demonstrated similar efficacy, the 200 mg sonidegib group was found to have longer duration of treatment, lower rate of adverse events, and lower discontinuation rate. The overall benefit-to-risk profile is more favorable with 200 mg than 800 mg sonidegib group. Quality of life was assessed based on two health questionnaires; the Quality of Life Questionnaires C30 (QLQ-C30) and the Head and Neck Cancer Module 35 (H&N35). The findings demonstrated the tolerability of sonidegib and improvement in or maintenance of quality of life in most patients with laBCC and mBCC.
Additional follow up of the BOLT clinical trial was ongoing at the time this review article was submitted.
Safety and tolerability
Both HPIs share similar adverse effects which include muscle spasms, dysgeusia (taste disturbance), alopecia, fatigue, nausea and weight loss. The vast majority of adverse effects were mild-to-moderate in severity. Treatment-related adverse effects were manageable and most were reversible after discontinuation of drug. The most frequent adverse events leading to discontinuation of HPI treatment were muscle spasms, dysgeusia and weight loss.
HPI-induced muscle contraction and muscle fiber twitching is thought to be the result of calcium influx [Teperino et al. 2012]. A study found amlodipine, a calcium channel blocker, reduced the frequency of vismodegib-induced muscle cramps [Ally et al. 2015]. This is the result of blocking voltage-gated calcium channels and inhibiting the transport of extracellular calcium into muscle which is required for contraction [Ozkul, 2007]. No clear relationship between the incidence of muscle cramps or spasm and elevated CK has been established [Migden et al. 2015]. Increased CK level is likely of skeletal muscle origin owing to the lack of evidence of cardiac muscle injury. Sonidegib is mainly metabolized by the hepatic enzyme CYP3A so drug interactions with CYP3A inhibitor can potentially increase the risk for further muscle toxicity [Zollinger et al. 2014].
Taste disturbances are well reported in patients treated with HPIs and some patients even discontinued the treatment because of this undesirable side effect [Dessinioti et al. 2014]. Chorda tympani nerve responses to taste stimuli were markedly reduced or absent in mice treated with LDE225 for up to 28 days [Kumari et al. 2015]. Vismodegib-treated mice were found to have reductions in taste bud size, taste cells per taste bud, and multiple taste-modulatory factors in the taste cells [Yang et al. 2015]. Regarding dysgeusia and weight loss, one study suggested the benefit of nutritional screening prior to initiation of vismodegib, and nutritional monitoring and management during the treatment [Le Moigne et al. 2016].
Alopecia is a well-known adverse effect of HPIs. The Hh pathway has an essential role during hair follicle morphogenesis, where it is required for normal advancement beyond the hair germ stage of development [Chiang et al. 1999]. Although HPI-induced alopecia appears to be reversible in most cases; 4 of 65 patients were reported to have persistent alopecia after stopping vismodegib [Alkeraye et al. 2015]. Death was reported in both HPI studies, however, a relationship between study drug and the deaths has not been established.
Drug resistance
Patients with treatment resistance to an SMO inhibitor may also fail other SMO inhibitors. A study of 9 patients with advanced BCCs that were previously resistant to treatment with vismodegib demonstrated resistance with sonidegib [Danial et al. 2015]. Genomic analysis of tumor biopsies revealed that vismodegib resistance is associated with Hh pathway reactivation, predominantly caused by SMO mutations, and to a lesser extent through mutation in suppressor of fused (SUFU), and gain-of-function mutations in Gli2 [Sharpe et al. 2015]. SMO mutations were identified in 50% (22 of 44) of resistant BCCs [Atwood et al. 2015]. The findings suggest that a combination of therapies targeting both SMO and downstream components of the Hh pathway may be required to overcome drug resistance.
Conclusion
HPIs provide a promising treatment for patients with advanced BCCs. Both vismodegib and sonidegib demonstrated similar overall efficacy, safety and tolerability. Vismodegib, the first HPI approved in the US, produced an ORR of 47.6% in patients with laBCC and 33.3% in patients with mBCC in the additional 12-month follow up. Sonigegib, the second HPI approved a few years later for laBCC, demonstrated an objective response in 36% of patients receiving 200 mg recommended dose. An additional 12 months of follow-up data was pending.
Future studies might include assessment of HPIs in combination with other treatment modalities to accomplish tumor control, and potential use of HPIs in advance of surgery to minimize potential postoperative disfigurement. Future effort is also needed to investigate the underlying mechanisms of actions for drug resistance.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Conflict of interest statement
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Michael Migden has received honoraria for advisory boards from Eli Lilly, Genentech, and Novartis. No other conflicts declared.