Hedgehog Signal Inhibition Hope or Hype?

Once a human malignancy metastasizes, currently available cytotoxic chemotherapy is usually palliative. While conventional chemotherapy regimens are associated with objective tumor regression, this only occurs in a minority of patients. Transient disease stabilization is common, but disease progression and ultimately death secondary to malignancy are expected. In most malignancies, conventional chemotherapy has reached a plateau in terms of response rate, progression-free survival, and most important, overall survival. While modest gains in response rate are achieved by combining a greater number of cytotoxic agents, increased toxicity is seen without improvement in survival. Recent discoveries in cellular and tumor biology have led to the development of novel targeted therapies that hold the promise of more selective anticancer treatment. One such molecular target is the Hedgehog signal transduction network.

The Hedgehog signal transduction pathway is named after a gene discovered by 1995 Nobel laureates Edward B. Lewis and Christiane Nüsslein- Volhard, who were screening for developmental genes causing body segmentation in Drosophila melanogaster (fruit fly). An experimental mutation in the Hedgehog gene caused fruit flies to develop as spikey balls reminiscent of a hedgehog, rather than with classic anterior-posterior segmentation. The Hedgehog signaling pathway plays a critical role in embryonic development, differentiation, and tissue polarity. Aberrant activation of this pathway occurs frequently in human cancer, whereas inactivation is associated with developmental disorders.

Examining Hedgehog Signaling

Hedgehog signaling is initiated by the binding of Hedgehog ligand to Patched, a 12-transmembrane protein receptor. This ligand-receptor family includes three known ligands, Sonic Hedgehog, Indian Hedgehog, and Desert Hedgehog, and two highly homologous receptors, Patched1 and Patched2. The Hedgehog pathway is repressed under normal conditions. In the absence of Hedgehog ligand, Patched constitutively inhibits Smoothened, a 7-transmembrane protein receptor that is a key downstream mediator of Hedgehog signaling. Downstream of Smoothened are the effector molecules of Hedgehog signaling: the Gli proteins (Gli1, Gli2, and Gli3), which function as zinc finger activators, repressors of transcription, or both. When Smoothened is inhibited, the Gli proteins are sequestered in the cytoplasm and bound to microtubules in a complex with the Fused (Fu) and Suppressor of Fused (SuFu) proteins; however, when Hedgehog binds Patched, the inhibition of Smoothened is relieved and the Gli proteins are released to enter the nucleus and induce transcription of Hedgehog target genes. There are a variety of downstream protein targets of Hedgehog signaling, including cyclins B1 and D1, p21, BCL-2, IL1R, and p53; these proteins are associated with mitosis, apoptosis, or resistance to therapy.

Implications of Patched Gene Mutations

Considerable additional insight into the role of the Hedgehog pathway in vertebrate development and human cancers has come from the discovery that mutations of the Patched gene are associated with a rare hereditary form of basal cell carcinoma (BCC): the Gorlin syndrome, also known as the basal cell nevus syndrome.^1,2 The most common manifestation of this disorder is the early development of multiple BCCs, usually on sun-exposed areas. Affected persons are also at risk for medulloblastomas, usually by age 3; palmar and/or plantar cutaneous pits; odontogenic cysts and other skeletal abnormalities; genitourinary anomalies, usually calcified ovarian fibromas; and a predisposition to unusual neoplasms such as sarcomas, cardiac fi bromas, and meningiomas.^3-6 Most germline mutations of Patched1 are loss- of-function nonsense mutations that lead to a constitutive activation of Smoothened. In sporadic BCCs, one copy of the Patched gene is frequently absent (loss of heterozygosity), while the other copy contains mutations, most of which are predicted to interrupt the function of Patched1.

Mice that are heterozygous for the Patched1 null mutation exhibit a phenotype that resembles the basal cell nevus syndrome, including the development of BCCs, medulloblastomas, and rhabdomyosarcomas. These observations indicate that Patched may function as a tumor suppressor. Activation of the Hedgehog signal transduction pathway due to loss-of-function mutations of Patched1 in human BCCs disrupts his feedback regulation, leading to uncontrolled Smoothened signaling. Activating mutations of Smoothened, on the other hand, are resistant to Patched1—mediated inhibition, which leads to an outcome similar to Patched1 inactivation. Endogenous activation of the Hedgehog pathway, through loss-of-function mutations of Patched, gain-of-function mutations of Smoothened, or loss-of-function of SuFu, is found in BCCs, trichoepitheliomas, medulloblastomas, and subsets of prostate cancer.

Hedgehog Pathway’s Role in Cancer

Since 2003, a growing body of data has indicated that the Hedgehog pathway is involved in the development, growth, and metastatic potential of a variety of cancers, including those of the lung, gastrointestinal tract, pancreas, and prostate. Unlike genetic mutations of Patched and Smoothened seen in BCCs and medulloblastomas, these tumors often appear to have elevated Hedgehog ligand expression. In some preclinical systems, such as BCC models, inhibition of Hedgehog signaling is associated with objective antitumor responses; in others, such as pancreatic cancer murine xenograft models, there is little effect against the primary tumor site, but there is a marked reduction in the number and sites of metastatic foci.

There is also evidence that Hedgehog signaling is potentially important in maintaining the tumor stem cell compartment.^7-9 For example, in the murine pancreatic cancer model, inhibition of Hedgehog signal transduction by the plant alkaloid cyclopamine reduced the number of putative pancreatic carcinoma stem cells and markedly reduced the number of subsequent metastatic tumor foci, although in that system there was little effect on the primary tumor itself.¹⁰

Hedgehog Inhibitors

In light of the poor outcome with current treatment for advanced cancer, new agents exploiting novel mechanisms must be evaluated. One such agent is GDC-0449, an orally bioavailable inhibitor of Hedgehog pathway signal transduction, for which the results of a phase I clinical trial have recently been published.¹¹ Of the 68 patients enrolled, 33 subjects with locally advanced or metastatic BCC were divided into three cohorts and given GDC- 0449 daily doses of 150 mg (n = 17), 270 mg (n = 15), and 540 mg (n = 1); these 33 patients formed the basis of the report, and no data were provided on non-BCC participants. The first dose was administered on day 1, followed by a second dose on day 8, with daily dosing onwards. This unique schedule allowed the collection of single- and multiple-dose pharmacokinetic data in the same patients. The half-life of the drug was 10 to 14 days. The steady-state serum levels were the same in all three dose cohorts, which indicates no need to increase the daily dose above 150 mg with this schedule. Skin punch biopsies and hair follicle assessments were used for pharmacodynamic analysis. Down-modulation of the Gli1 transcription factor was observed in all skin punch biopsy samples after treatment with the drug. GDC-0449 was extremely well tolerated, and no dose-related toxicities were observed. Adverse events considered to be possibly or probably drug-related included grade 3 fatigue in four patients; grade 1-2 dysguesia in three patients; and grade 3 hyponatremia, muscle spasm, and atrial fibrillation in one patient each. One episode of grade 4 hyponatremia was thought to be unrelated to the study drug.

Of the patients with BCCs, partial response was seen in 18, with stable disease observed in 11 and disease progression noted in the remaining four. Median progression-free survival was longer than 8.8 months at the time of publication. One additional patient who was referred for participation in the phase I trial was included in a separate report by Rudin and colleagues.12 This patient was a 26-year- old man with widely metastatic and refractory medulloblastoma that demonstrated biochemical evidence of Hedgehog pathway activity. He was treated off protocol because of pancytopenia, presumed secondary to bone marrow infiltration. The patient experienced a marked clinical partial response to GDC-0449 at a total daily dose of 540 mg by the second month, evidenced by relief of symptoms and by physical and radiographic examination; however, 1 month later, radiographic evaluation showed disease progression, and he was taken off therapy. The patient died 2 months later. While the response was brief, this case report highlights a fundamental principle in the development of Hedgehog and other targeted therapies: cancers selected for target activation are more likely to be susceptible to target inhibition. The key challenge is identification of valid measurements of pathway dependence.

Multiple phase II studies involving GDC-0449 are now recruiting. Examples include GDC-0449 versus placebo in combination with standard chemotherapy (5-fl uorouracil, folinic acid, oxaliplatin, and bevacizumab) for first-line treatment of metastatic colorectal cancer, and first-line therapy for small- cell lung cancer in combination with etoposide and cisplatin. In addition, studies are being planned in advanced BCC, as well as maintenance therapy after a second or third remission from conventional chemotherapy in patients with ovarian adenocarcinomas.

While GDC-0449 has demonstrated clinical safety and potential activity, additional agents targeting the Hedgehog signaling pathway are beginning to enter clinical trials. Some agents are potentially limited in their applicability. For example, Hedgehog-binding antibodies, similar to the vascular endothelial growth factor—binding bevacizumab (Avastin), will be ineffective if Hedgehog pathway activation results not from overexpression of ligand, but instead from an abnormality of Patched, Smoothened, or effector molecules further downstream. An additional agent, such as forskolin, a protein kinase A inhibitor, would be active in blocking activity downstream but would be associated with significant potential adverse effects due to the ubiquitous nature of protein kinase A signaling.¹³ However, in addition to cyclopamine, which is too toxic for human use, a variety of naturally occurring compounds have demonstrated the ability to inhibit signal transduction through the Hedgehog pathway.¹⁴ Other agents, such as IPI-926 and XL-139 (BMS-833923), are also now in phase I clinical trials as single agents and in combination with standard cytotoxic drugs, and will likely rapidly progress into phase II evaluation in the absence of major toxicities.^15-16

Putting Cancer Treatments in Perspective

Cytotoxic chemotherapy has produced substantial palliation in a majority of advanced adult malignancies and, in the setting of adjuvant therapy, has undoubtedly contributed to curing some cancers. The level of benefit with currently available agents, however, has apparently reached a plateau that has not yielded dramatically to variations in dose, schedule, combination, or number of concurrent drugs. Recent appreciation of the molecular pathways upon which malignant cells depend has allowed the development of novel treatment strategies that have produced clinical benefit for a substantial number of patients. A note of caution is in order, however: as with most targeted therapies, particularly the small-molecule signal transduction inhibitors, it is unlikely that a single agent will provide substantial clinical benefit for a majority of patients with a single disease, or across a broad spectrum of diseases. This is a result largely of the genetic heterogeneity of most malignancies, such that only a small fraction truly depends upon a single signal transduction pathway. Examples include the Philadelphia chromosome in chronic myelogenous leukemia, EGFR-exon 19— or EGFR-exon 21–mutated non-small cell lung cancer, and Hedgehog-dependent BCC. While dramatic results can be seen in small patient populations in relatively uncommon settings (eg, imatinib in chronic myelogenous leukemia), such responses will be unlikely in unselected patients with more common malignancies such as those of the colon, breast, prostate, or lung. As for most other targeted small-molecule inhibitors, Hedgehog signal inhibitors are likely to be effective in cancers selected for evidence of pathway activation. This is an unsurprising manifestation of the paradigm established by the first targeted therapy, estrogen manipulation in breast cancer. At present, it would be unthinkable to offer tamoxifen or an aromatase inhibitor to a woman whose breast cancer showed no evidence of either the estrogen or progesterone receptor.

Take-home message

If the patient population most likely to benefit from Hedgehog inhibitor therapy could be identified, then the development of such drugs would provide hope, not empty hype. Further studies will likely identify predictors of clinical benefit and determine who among patients with more common cancers may benefit from Hedgehog signaling blockade or other targeted therapies. The size of candidate populations may number only in the thousands, but the unmet need is great, and the benefit to that minority is likely to be very large. The era of finding blockbusters has largely passed, and clinical trials of targeted agents are ushering in a new era of truly individualized therapies.

Julie E. Bauman, MD, and Dennie V. Jones, Jr, MD, practice in the Department of Internal Medicine/ Division of Hematology and oncology, University of New Mexico Cancer Center, Albuquerque, New Mexico.

REFERENCES

1. Hahn H, Wicking C, Zaphiropoulous PG, et al. Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell. 1996;85(6):841- 851. Available at: www.ncbi.nlm.nih.gov/pubmed/8681379.

2. Johnson RL, Rothman AL, Xie J, et al. Human homolog of patched, a candidate gene for the basal cell nevus syndrome. Science. 1996;272(5268):1668-1671. Available at: www.ncbi.nlm.nih.gov/pubmed/8658145.

3. Evans DG, Ladusans EJ, Rimmer S, et al. Complications of the naevoid basal cell carcinoma syndrome: results of a population

based study. J Med Genet. 1993;30(6):460- 464. Available at: http://jmg.bmj.com/content/30/6/460.abstract.

4. Veenstra-Knol HE, Scheewe JH, van der Vlist GJ, et al. Early recognition of basal cell naevus syndrome. Eur J Pediatr. 2005;164(3):126-130. Available at: http://tinyurl.com/yd79mjl.

5. Kimonis VE, Goldstein AM, Pastakia B, et al. Clinical manifestations in 105 persons with nevoid basal cell carcinoma syndrome. Am J Med Genet. 1997;69(3):299-308. Available at:http://tinyurl.com/ykwaqw3.

6. Gorlin RJ. Nevoid basal-cell carcinoma syndrome. Medicine (Baltimore). 1987;66(2):98-113. Available at: www.ncbi.nlm.nih.gov/pubmed/3547011.

7. Katoh Y, Katoh M. Hedgehog signaling pathway and gastrointestinal stem cell signaling network. Int J Mol Med. 2006;18(6):1019-1023. Available at: http://tinyurl.com/yg4bcuz.

8. Zhou JX, Jia LW, Liu WM, et al. Role of sonic hedgehog in maintaining a pool of proliferating stem cells in the human fetal epidermis. Hum Reprod. 2006;21(7):1698- 1704. Available at: http://tinyurl.com/yjgcwjp.

9. Fu JR, Liu WL, Zhou JF, et al. Sonic hedgehog protein promotes bone marrow- derived endothelial progenitor cell proliferation, migration and VEGF production via PI 3-kinase/Akt signaling pathways. Acta Pharmacol Sin. 2006;27(6):685-693. Available at: http://tinyurl.com/yz9nkc9.

10. Feldmann G, Dhara S, Fendrich V, et al. Blockade of hedgehog signaling inhibits pancreatic cancer invasion and metastases: a new paradigm for combination therapy in solid cancers. Cancer Res. 2007;67(5):2187-2196. Available at: http://tinyurl.com/yzvwhts.

11. Von Hoff DD, LoRusso PM, Rudin CM, et al. Inhibition of the hedgehog pathway in advanced basal-cell carcinoma. N Engl J Med. 2009;361(12):1164-1172. Available at: http://content.nejm.org/cgi/content/short/361/12/1164.

12. Rudin CM, Hann CL, Laterra J, et al. Treatment of medulloblastoma with hedgehog pathway inhibitor GDC-0449. N Engl J Med. 2009;361(12):1173-1178. Available at: http://tinyurl.com/ybng84t.

13. Stecca B, Ruiz i Altaba A. The therapeutic potential of modulators of the Hedgehog- Gli signaling pathway. J Biol. 2002;1(2):9. Available at: http://tinyurl.com/yecax4j.

14. Hosoya T, Arai MA, Koyano T, et al. Naturally occurring small-molecule inhibitors of Hedgehog/GLI-mediated transcription. Chembiochem. 2008;9(7):1082-1092. Available at: http://tinyurl.com/yclynmc.

15. Tremblay MR, Lescarbeau A, Grogan MJ, et al. Discovery of a potent and orally active hedgehog pathway antagonist (IPI-926). J Med Chem. 2009;52(14):4400-4418. Available at: http://tinyurl.com/ykj3d4w.

16. Olive KP, Jacobetz MA, Davidson CJ, et al. Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science. 2009;324(5933):1457-1461. Available at: www.sciencemag.org/cgi/content/abstract/324/5933/1457.