Hedgehog inhibitor gets landmark skin cancer approval, but questions remain for wider potential

Approval for Genentech's vismodegib marked the first approval for a Hedgehog inhibitor, but questions remain for the class's broader role in other cancers. You have full access to

this article via your institution. Download PDF In January, Genentech's vismodegib became the first drug to be approved by the US Food and Drug Administration for advanced basal cell

carcinoma (BCC), which is positive news for these patients who have had few treatment options. Perhaps just as important as the clinical milestone, vismodegib's approval represents a

significant scientific achievement: the drug is the first on the market to target the Hedgehog (HH) signalling pathway. > We have been dreaming of this for 15 years. “We are overjoyed

with the news that a HH signalling inhibitor is at last on the market — we have been dreaming of this for 15 years,” says Jingwu Xie of Indiana University, Indianapolis, USA, who was part of

one of the teams that in 1996 identified the mutated gene responsible for BCC. Yet although vismodegib's approval for metastatic and inoperable locally advanced BCC is encouraging news

for the field, an elephant in the room looms large: will HH inhibitors prove to be effective in a broader range of more common tumour types? Vismodegib and other HH inhibitors act as

targeted therapies in the BCC setting (in which specific HH pathway mutations lead to aberrant pathway activity), but despite implications that HH signalling is important in other cancers

there is no clear genetic basis for efficacy in these. “In BCC, HH signalling is the cause of the disease, whereas in other cancers its activation may be only one of many drivers,” explains

Xie. “We probably don't understand yet how to use [HH inhibitors] in these diseases,” adds Genentech's Frederic de Sauvage, who initiated biological studies of the HH pathway at

the company in 1996. Vismodegib is also the first of a new broader class of anticancer agents that target embryonic signalling pathways, which also include Notch and WNT. Researchers

commented, however, that because each of these operate independently — with distinct targets and biomarkers, and quite often in different tumour types — there are few lessons from HH

development that can be extrapolated. HOMING IN ON HEDGEHOG The HH pathway was first identified in 1980 in the fruitfly, when researchers identified mutations in a gene that caused the

insects to develop abnormally; the flies looked spiky, and so scientists named the gene and pathway 'Hedgehog'. A related pathway in humans similarly has a crucial role in

development, modulating cell growth and differentiation at the cellular level, with a particular importance within the neural tube and skeleton. In most adult tissues it is silenced. In

1996, direct evidence of a cancer link came to light when different teams of researchers identified the mutated gene responsible for Gorlin syndrome: a rare, inherited condition that

predisposes patients to BCC. This gene, patched 1 (_PTCH1_), encodes a transmembrane receptor for several HH protein ligands. Soon after the link to hereditary BCC was elucidated,

researchers independently implicated abnormal HH signalling in over 90% of cases of sporadic BCCs. In animal models, activation of the pathway was shown to be sufficient to drive BCC tumour

formation. The first HH pathway inhibitor was discovered long before the pathway. After finding that sheep that ate the poisonous plant _Veratrum californicum_ gave birth to one-eyed lambs,

a phenotype called cyclopia, researchers set out to understand the cause. In 1968 they identified the culprit, a compound they named cyclopamine. Nearly three decades later, cyclopamine was

found to block HH activity by binding the Smoothened receptor (SMO), a transmembrane protein whose function is normally inhibited by PTCH1 in the absence of HH ligand. When HH ligand binds

to PTCH1, SMO is released to signal downstream, especially via the downstream GLI transcription factors. Vismodegib is also a SMO inhibitor, but with greater potency and more favourable

pharmaceutical properties than cyclopamine. So far, all of the other HH inhibitors in clinical development are also SMO inhibitors (Table 1), although they are structurally distinct from one

another. Many researchers would like to be able to block GLI activity, to overcome resistance to SMO inhibition through SMO mutation (which can happen in BCC), and for cases in which GLI is

activated through independent oncogenic signalling pathways, but have so far only had preliminary preclinical success. CLINICAL QUESTIONS REMAIN In the 104-patient single-arm Phase II study

of vismodegib that enabled approval, the drug resulted in a 43% response rate in patients with locally advanced disease, and a 30% response rate in patients with metastatic disease. Median

progression-free survival for both groups was 9.5 months, but overall survival data are not yet available. Historically, median overall survival has been reported to be about 8 months for

metastatic BCC. In other indications, however, the clinical data to date have not been promising. Development of vismodegib in metastatic colorectal cancer and in advanced recurrent ovarian

cancer was suspended in 2010 after Phase II trials failed to show sufficient efficacy. Infinity's saridegib, one of the other leading HH inhibitors, failed earlier this year to show an

overall survival benefit in metastatic pancreatic cancer. These failures stem largely from an incomplete understanding of how HH signalling may contribute to disease in these cancers.

Whereas mutations in _PTCH1_ and _SMO_ lead to aberrant HH pathway activation in BCC (and in some rare childhood medulloblastomas), they are not routinely found in other tumour types such as

colorectal cancer, ovarian cancer and pancreatic cancer. HH ligand overexpression, via an unspecific mechanism, is instead thought by some researchers to sometimes lead to aberrant pathway

activity. “HH has been implicated in many tumours, but the jury is still out on whether there is a role for HH inhibitors in cancers where the pathway is not mutated,” says de Sauvage. In

addition, it is unclear whether HH signalling acts oncogenically on cell proliferation, survival and/or metastasis, and whether it acts via paracrine or autocrine mechanisms in different

tumour types. “If we don't know the real mechanism, it is hard to target the pathway,” says Xie. Clinical trial data have as yet provided little insight into these questions, but

William Matsui of Johns Hopkins University School of Medicine, Baltimore, Maryland, USA, hopes they will soon. Xie, meanwhile, argues for more careful preclinical studies using orthotopic

rather than subcutaneous mouse models. Another key possibility for the HH pathway is that, along with the WNT and Notch pathways, it is important for the regulation of the 'cancer stem

cells' (CSCs), a small subset of tumour cells that are believed to be the cause of resistance to many cancer therapies and lead to disease recurrence. HH–GLI1 signalling may have a

crucial role in promoting CSC self-renewal and progenitor expansion across many tumour types, says Ariel Ruiz i Altaba, of the University of Geneva, Switzerland, who led a team that linked

sporadic BCC and HH–GLI1 signalling in the 1990s. However, the primary data supporting this theory come from only a small set of tumour types, and there are extremely limited data from

humans to back it up. “There are more claims from limited experiments than evidence to support the theory that HH works through regulating CSCs,” says Xie. Further studies, including plans

to measure the numbers of CSCs before and after treatment in US National Cancer Institute trials of vismodegib, are underway to provide further clarity. > To know which tumours may 

respond, we need to better understand the > details. Another issue is that the field currently only understands the skeleton of the pathway, says Ruiz i Altaba. “To know which tumours may

respond, we need to better understand the details.” There is also confusion about how common HH signalling activation is in different cancers. This may be due to a lack of samples analysed

of a certain tumour type, because HH signalling may only be active in the hard-to-detect CSCs, or because different researchers have used different standards to define pathway activation.

“There was a group that showed that almost 100% of breast cancers had HH pathway activation, whereas another showed almost none had this,” says Xie. Many researchers, therefore, are hunting

for biomarkers that can robustly show pathway activation, such as expression levels of HH ligand, SMO and GLI1, to predict patient response. If paracrine signalling is involved, downstream

stromal biomarkers such as WNT, insulin-like growth factor and vascular endothelial growth factor may be important, adds de Sauvage. As yet, however, no robust biomarkers — or evidence of

benefit in subgroup populations — have been reported from clinical trials. Through whichever mechanisms HH signalling may be working, researchers generally do not believe that — even if it

is a driver pathway — it is likely to be the only one. Understanding the crosstalk between it and other tumour-promoting pathways is therefore crucial to uncovering potential benefit from

combination therapies. Vismodegib's approval may spur combination testing, given that “once a drug gets approved, companies are more comfortable testing it in combination with other

drugs”, says Glen Weiss at TGen in Phoenix, Arizona, USA. Possible combination candidates include inhibitors of well-known oncogenic pathways, such as RAS, which may cause GLI1 activation in

a SMO-independent manner, or WNT or Notch inhibitors, in cases where these are co-activated. “We don't know yet where HH inhibitors will be most effective — the field is wide open,”

concludes Matsui. Authors * Malini Guha View author publications You can also search for this author inPubMed Google Scholar RIGHTS AND PERMISSIONS Reprints and permissions ABOUT THIS

ARTICLE CITE THIS ARTICLE Guha, M. Hedgehog inhibitor gets landmark skin cancer approval, but questions remain for wider potential. _Nat Rev Drug Discov_ 11, 257–258 (2012).

https://doi.org/10.1038/nrd3714 Download citation * Published: 30 March 2012 * Issue Date: April 2012 * DOI: https://doi.org/10.1038/nrd3714 SHARE THIS ARTICLE Anyone you share the following

link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature

SharedIt content-sharing initiative