The Hedgehog pathway is one of several key developmental cell-signaling networks whose aberrant activation in adult tissues can lead to cancer.
Named for its spiky appearance in mutated fruit fly embryos, the Hedgehog pathway is one of several key developmental cell-signaling networks whose aberrant activation in adult tissues can lead to cancer.
Thus far, more than 30 years of exploration have yielded two small-molecule inhibitors for the treatment of patients with metastatic or locally advanced basal cell carcinoma (BCC). The FDA approved vismodegib (Erivedge) in January 2012 and sonidegib (Odomzo) in July 2015, both for indications in recurrent disease after surgery and radiotherapy.
The successful development of these two agents has validated the Hedgehog pathway as a bona fide therapeutic target, and clinical trials are under way to determine whether similar inhibition strategies can be employed against other tumor types including breast, prostate, brain, and gastrointestinal cancers. Thus far, the most promising target in the pathway has proved to be the Smoothened (Smo) receptor (Table).
aTrial is ongoing, but not recruiting participants.
bMonoclonal antibody that targets insulin-like growth factor-1 receptor. AML indicates acute myeloid leukemia; BCC, basal cell carcinoma; CML, chronic myeloid leukemia; CMML, chronic myelomonocytic leukemia; JAK, Janus kinase; MDS, myelodysplastic syndrome; RT, radiation therapy; SCLC, small cell lung cancer; TNBC, triple-negative breast cancer.
At the same time, researchers have acknowledged many deficits in our understanding of this highly complex signaling network including the importance of delineating mechanisms of resistance to currently available therapies.The key players in the Hedgehog pathway are the transmembrane receptor Patched (Ptch), its three ligands (Sonic, Indian, and Desert Hedgehog); Smo, a second receptor that is structurally similar to G-protein—coupled receptors; and a family of three glioma-associated transcription factors (Gli1, Gli2, and Gli3).
In mammals, canonical Hedgehog signaling occurs predominantly in the primary cilium, an antenna-like structure that protrudes from the surface of most cells. In its “off” state, when no ligand is present, Ptch inhibits Smo by mechanisms that are not yet clear. In the “on” state, ligand binding to Ptch causes it to release its inhibition on Smo, which is then free to move into the primary cilium and propagate the signal downstream.
This culminates in the activation of the Gli transcription factors, which regulate the activity of Hedgehog target genes that control vital cellular processes necessary for embryonic development, including cell proliferation, migration, metabolism, and determination of cell fate.
Hedgehog ligands are morphogens, produced by specialized secreting cells and, acting in either an autocrine (on the same cell) or paracrine (on a different cell) fashion; their concentration at the target cell determines the response that is elicited from that cell. This graded response is achieved in part by impacting the proteolytic processing of the Gli proteins. Depending on the strength of the Hedgehog signal, different ratios of full-length active Gli isoforms to truncated suppressive isoforms are generated and regulate the strength of the transcriptional response.
The steps between Smo and Gli are poorly defined, but one protein that has been identified is Suppressor of Fused (SuFu), which plays an important role in the graded response to Hedgehog signals by mediating processing and degradation of the Gli proteins.
Although this description has become the established view of Hedgehog signaling, there is a growing body of evidence that the pathway can be activated in other ways. In so-called noncanonical signaling, the components of the pathway may be able to regulate basic cellular processes seemingly independently of the full pathway. These routes of activation may have a particularly important role in the development of cancer.Although the Hedgehog pathway plays a critical role in embryonic development, the network is mostly inactive in adults, though it does have important functions in regenerating tissue and in regulating the growth of stem cells, particularly in the brain.
Since the signaling network controls the expression of a number of genes involved in cancer hallmark processes, inappropriate activation of the pathway is found in a variety of different tumor types. Oncogenic activation of Hedgehog signaling is thought to come about via several different mechanisms:
The type of activation is linked to the specific tumor type, such that ligand-independent activation is more often associated with BCC and medulloblastoma. Indeed, the first suggestion of a role for Hedgehog signaling in cancer stemmed from the discovery that Gorlin syndrome, a condition in which people are prone to the development of numerous BCCs and other forms of cancer, is driven by an inherited mutation in PTCH1.
Subsequently, PTCH1 mutations have been identified in more than 90% of patients with BCC and approximately 20% to 30% of medulloblastoma cases. Mutations in SMO, SUFU, SHH, and GLI1/2 have also been observed in BCC and many other tumor types. Overall, inappropriate Hedgehog signaling has been demonstrated in more than 30% of human cancers, including breast, ovarian, pancreatic, cervical, and lung cancers.To date, the most extensively studied Hedgehog- targeting agents have been small-molecule inhibitors targeting the Smo protein. Development of these drugs began with the discovery that cyclopamine, a compound derived from the California corn lily and a naturally occurring Smo inhibitor, affected tumor growth in preclinical trials. Poor pharmacokinetics impeded its clinical development, but it served as the stimulus for screens to identify small molecule synthetic derivatives.Vismodegib, which targets Smo, became the first FDA-approved Hedgehog inhibitor after demonstrating overall response rates (ORRs) of 30.3% and 42.9% among patients with metastatic BCC and locally advanced BCC, respectively, who participated in a single-arm clinical trial. The participants with metastatic disease who responded to vismodegib all experienced partial responses (PRs), while 13 of 63 evaluable patients (20.6%) with locally advanced BCC achieved complete responses (CRs). The most common adverse events of all grades, evaluated among patients who participated in four clinical trials, were muscle spasms (71.7%), alopecia (63.8%), and dysguesia (55.1%). The most prevalent grade 3 adverse events included weight loss (7.2%), fatigue (5.1%), and muscle spasms (3.6%).In July, vismodegib was joined by sonidegib, an Smo inhibitor that was approved based on the results of the phase II BOLT trial. In this noncomparative study, 230 patients were randomized 2:1 to two different regimens of sonidegib (200 mg or 800 mg daily). The FDA based its approval on an analysis of 66 patients with locally advanced BCC, in whom an ORR of 58% was observed at a dose of 200 mg, including three CRs and 35 PRs. Patients who received sonidegib also experienced musculoskeletal adverse reactions including spasms, pain, and myalgia. The most serious risks were rhabdomyolysis and embryo-fetal toxicity.
Illustrations courtesy of Genentech.
These illustrations depict downstream elements of the Hedgehog pathway and the juncture at which vismodegib (Erivedge), the first approved drug to target the network, interferes with oncogenic signaling. At left, carcinogenic activity occurs when the activated transmembrane protein Smoothened triggers glioma-associated transcription factors that regulate vital cell processes. Vismodegib, right, binds to and inhibits Smoothened.
In an interview with OncologyLive, Anne Lynn S. Chang, MD, director of the Advanced Basal Cell Carcinoma Clinic at Stanford School of Medicine, highlighted the importance of this approval. “Clearly, more choices are better,” she said. “A patient may be intolerant of vismodegib because of the side effects, so to have another option, a different molecule, may prove a better treatment choice for that particular patient.”Although these two approvals represent a substantial advancement for the treatment of BCC, attempts to employ Hedgehog inhibitors against other tumor types have not yet produced results. Medulloblastoma is considered a rational target for Smo inhibitors because Gorlin syndrome also commonly results in the development of these brain tumors. Some efficacy has been noted, including two recent phase II Pediatric Brain Cancer Consortium studies, PBTC-025B and PBTC-032, in pediatric and adult patients with recurrent medulloblastoma. Responses were observed in three adult patients and one child, all of whom had tumors with Sonic Hedgehog (SHH) mutations, and prolonged disease stabilization was observed in more than 40% of patients with the mutation.
To date, however, the use of Smo inhibitors in other solid tumors has proved more miss than hit. For example, in patients with ovarian cancer, disappointing results from a phase II study were reported in 2012 that suggested vismodegib did not improve progression-free survival (PFS) when used as maintenance therapy in patients in second or third complete remission, but results were confounded by lower than expected levels of Hedgehog ligand expression. More recently, Smo inhibitors have shown promising activity in hematologic malignancies.
In a study published in Lancet Haematology, PF-04449913 demonstrated a suggestion of clinical efficacy in 49% of patients with hematologic malignancies treated with doses ranging from 5 mg to 600 mg per day for up to 12 cycles of 28 days. Among 28 patients with acute myeloid leukemia, 16 patients responded, including one complete remission and four partial remissions.The most logical strategy for improving the efficacy of Smo inhibitors is to combine them with other effective therapies. Such regimens already are being explored in clinical trials with vismodegib or sonidegib. Preliminary data on two patients with BCC treated with concurrent vismodegib and radiation therapy suggest that the combination may be effective and well tolerated.
In other cancers, combination therapy seems less effective. A phase II study in patients with colorectal cancer, in which vismodegib was combined with chemotherapy as firstline treatment for metastatic disease, reported that the combination did not add to the efficacy of standard therapy while increasing toxicity.
In a pilot study of vismodegib plus gemcitabine in patients with metastatic pancreatic adenocarcinoma, the combination was not superior to gemcitabine alone. However, preliminary data from an ongoing single-arm phase II trial of vismodegib in combination with gemcitabine and nab-paclitaxel was reported at the 2014 Gastrointestinal Cancers Symposium. More than 80% of 49 evaluable patients had stable disease or better, including one CR and 20 PRs. In the overall study population of 59 patients, the median PFS was 5.5 months and overall survival was 10 months.Resistance is a common foe of targeted cancer therapies and Hedgehog-targeted agents have not been spared this fate. “Around 50% of patients don’t respond to monotherapy with vismodegib or sonidegib,” said Chang. Furthermore, tumors that initially respond to Smo inhibitors can acquire resistance over the course of treatment. Chang noted that in her practice they have found that this happens at 9 months in approximately 20% of patients treated with vismodegib,and published those findings.
According to gene expression analyses of resistant tumors, resistance is often linked to mutations in SMO that impact drug binding (eg, a D473H variant is commonly associated with acquired resistance), but also to dysregulation of other components of the Hedgehog pathway that activate signaling in the absence of Smo, or even upregulation of other signaling pathways, such as the phosphatidylinositol-3-kinase (PI3K) pathway. In an effort to combat resistance, inhibitors that can bind mutant Smo are being examined. Taladegib (LY2940680) is in phase I clinical trials and has demonstrated anticancer activity in D473H-mutant cell lines.
Combination with other targeted therapies could also help; Chang highlighted an ongoing clinical trial of sonidegib with a PI3K inhibitor, based on the observation of PI3K pathway activation in some resistant tumor samples. At this point, Chang said, it is too early to deduce how this may impact response rates.
Another possible solution for overcoming resistance is to develop drugs targeting other proteins in the Hedgehog pathway that are downstream of Smo. The most promising alternative target is the Gli transcription factors. Several agents that are thought to inhibit these proteins have been described, including the antifungal agent triazole itraconazole, natural compounds such as staurosporine and physalin B, and the drug arsenic trioxide, which is currently used for the treatment of acute promyelocytic leukemia. Synthetic Gli-targeting agents have also been developed, such as the smallmolecule inhibitors Gant58 and Gant61, but none of these drugs has progressed beyond preclinical testing as yet.
Jane de Lartigue, PhD, is a freelance medical writer and editor based in New Haven, Connecticut.