Molecular mechanisms and clinical use of targeted anticancer drugs

The last decade has seen the introduction of several new classes of targeted anticancer therapies for routine clinical use.

Unlike chemotherapy, which generally targets all dividing cells, these drugs are specific for a molecular characteristic of a cancer such as a growth factor receptor or signalling molecule.

Although targeted therapies do not cause the antiproliferative toxicities typical of chemotherapy, they do have adverse effects of their own.

These new drugs include monoclonal antibodies, such as bevacizumab and rituximab, and the small molecule tyrosine kinase inhibitors, such as dasatinib and sorafenib.

Targeted therapies are often taken for long periods of time. Many of the drugs, such as the tyrosine kinase inhibitors, are orally administered so patients receiving these therapies are increasingly being seen in general practice.

Historically, the treatment of cancer was based on an understanding of the differences in cell kinetics between normal and malignant cells. The largely fortuitous discovery of cytotoxic therapy resulted in a class of drugs that has become known as chemotherapy. Most chemotherapy drugs have a direct action on DNA or proteins involved in DNA replication, translation and transcription. Although it was hoped that chemotherapy would selectively treat the disease and not normal tissue, this was not the case. As chemotherapy generally targets all dividing cells, highly proliferative normal tissues are also affected and common toxicities develop such as myelosuppression and mucositis.

The search for more targeted cancer therapies has been supported by a better understanding of malignancy at the molecular level. Cancers exhibit acquired characteristics that enable their malignant phenotype. These have been called the hallmarks of cancer and traits include:

• replicative immortality
• sustained proliferative signalling
• evasion of growth suppressors
• angiogenesis induction
• tissue invasion and metastases
• resistance to cell death
• deregulated cell metabolism
• genomic instability
• tumour-promoting inflammation
• ability to evade the immune system.

Cancers interact with their local microenvironment or stroma through angiogenesis, inflammation and immune responses. Each of these hallmarks of cancer provides a target for drug therapy.

The targeted therapies can be broadly divided into two classes:

• monoclonal antibodies (Table 1)
• small molecules – predominantly tyrosine kinase inhibitors (Table 2).

Monoclonal antibodies are denoted by the suffix –mab, for example trastuzumab for breast cancer. They typically require intravenous or subcutaneous administration. These antibodies are produced by recombinant DNA technology and may consist of human and non-human protein, or be partially or fully humanised. Chimeric antibodies* are more likely to elicit hypersensitivity reactions due to pre-existing immunity to foreign animal protein.

Monoclonal antibodies target cell surface molecules, usually receptors, or their ligands. They may exert their effects through interference with a receptor pathway or through immune mechanisms such as antibody-dependent cellular cytotoxicity.

The small molecules typically block pathways that are continuously activated in cancer cells. The tyrosine kinase inhibitors are the most common and work by inhibiting kinases that phosphorylate key proteins to activate signal transduction pathways. They are denoted by the suffix –nib, for example imatinib for chronic myeloid leukaemia, and are typically developed for oral administration. These inhibitors block a number of different tyrosine kinases.

Another class of small molecules is the inhibitors of mammalian target of rapamycin (mTOR). They have the suffix –imus, for example everolimus for pancreatic neuroendocrine tumour or temsirolimus for renal cell carcinoma. These molecules bind to an intracellular protein (FKBP-12). This complex then blocks the activity of mTOR kinase which inhibits angiogenesis and tumour cell growth, proliferation and survival.

Targeted therapies are developed on the premise that a particular target is important in the pathogenesis of a malignancy. The relevance of a target may have been determined through basic scientific research or through epidemiological studies in patients with tumours that express the target. Often targets are prognostic biomarkers. For example, human epidermal growth factor receptor 2 (HER2)/neu amplification is associated with poor prognosis in breast cancer. Also, the KRAS (Kirsten rat sarcoma-2 virus oncogene) mutation is associated with poor prognosis in colon cancer.

Ideally, targeted therapies have very high specificity for their target. If the tumour does not express the target then the therapy will be ineffective. Target molecules are predictive biomarkers for efficacy of the therapy. For example, trastuzumab is ineffective in treating breast tumours that do not have amplified HER2/neu. On the other hand, cetuximab is only effective for colorectal cancers that have wild-type (normal) KRAS, but not tumours that have mutant KRAS (Table 1).

As targeted therapies are dependent on their target it should be possible to tailor therapy to suit individual patients. This therapeutic strategy avoids treating patients who will not benefit and may only experience adverse effects. In addition more selective use of the therapy results in improved cost-effectiveness.

In Australia, prescribing of many targeted therapies requires evidence that the patient's cancer is expressing the target molecule. This is confirmed using a companion diagnostic test. Future government registration and reimbursement of targeted therapies will include parallel assessment of any associated diagnostic test.

In general, targeted therapies do not cause the antiproliferative toxicities of chemotherapy, but they all have toxicities of their own (Tables 1 and 2). Some of the toxicities are common to the classes of drug. For example:

• acneiform rash with drugs that target the epidermal growth factor receptor
• cardiac toxicity with drugs targeting HER2/neu
• hypertension and other vascular toxicities with angiogenesis inhibitors such as bevacizumab.

Some of the small molecules have multiple targets and have a greater potential to cause adverse effects than monoclonal antibodies. Angiogenesis inhibitors such as bevacizumab may interact with a patient's pre-existing medical condition, for example hypertension.

Patients who were not previously suitable for chemotherapy may now receive treatment with one of the new drugs. Targeted therapies are also more suitable as maintenance therapies than cytotoxics.

Response to traditional chemotherapy is usually determined by a change in tumour size. While this is still possible with targeted therapies, many of the drugs stabilise tumours rather than shrink them. Sometimes these drugs are referred to as cytostatic rather than cytotoxic. An example is sorafenib for hepatocellular carcinoma. Fewer than 2% of patients achieve a partial response, but their time to progressive disease is significantly longer than with placebo. This leads to improved survival.

Drugs such as imatinib for gastrointestinal stromal tumours can alter radiological appearance on CT scan, making tumours more hypodense compared to baseline. Specific response criteria, called the Choi criteria, use change in the appearance of the tumour, rather than change in size, as one of the response measures.¹ Positron emission tomography may show gastrointestinal stromal tumours have also become hypometabolic after a short period of treatment despite no objective change in tumour size.

Sometimes it is difficult to measure a patient's response to therapy and newer measures of treatment response are being developed. For example, patients receiving immunotherapy with the anti-cytotoxic T-lymphocyte associated antigen 4 (CTLA-4) monoclonal antibody ipilimumab for metastatic melanoma may have objective evidence of disease progression before an immune response occurs. The immune-related response criteria have been developed to take this observation into account.²

All of the common cancers and many rare tumours now have at least one line of systemic therapy. GPs will routinely encounter patients receiving a targeted therapy and it will be useful to be aware of common toxicities (Tables 1 and 2). Perhaps more important is the role the doctor can play in ensuring treatment compliance. Many targeted therapies are given continuously, for example imatinib for gastrointestinal stromal tumour and erlotinib for non-small cell lung cancer. If treatment is interrupted, the patient's disease may progress. Although the tumour may respond when rechallenged, resistance to therapy may emerge. As such the GP can encourage patient adherence and also liaise with the cancer specialist before stopping therapy because of toxicity.

It is expected that many more targeted therapies will come into routine clinical use. A future direction for small molecule tyrosine kinase inhibitors will be to combine them to overcome treatment resistance. Monoclonal antibodies will be modified to become carriers for radiation or cytotoxic drugs and will be enhanced to increase their immune effects. The use of these medicines will be improved by further development of companion diagnostic tests.

Winston Liauw has received funding from Merck-Serono and has ongoing participation as an investigator in numerous clinical trials associated with the development of targeted drugs. He provides drug development and medical advice to Clinical Network Services. He serves on the board of directors of NPS MedicineWise.

* antibodies that contain polypeptides from different species

Piccart-Gebhart MJ, Procter M, Untch M, Smith I, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005;353:1659-72.

Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335-42.

Karapetis CS, Khambata-Ford S, Jonker DJ, O'Callaghan CJ, Tu D, Tebbutt NC, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008;359:1757-65.

Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646-74.

Cancer Institute NSW. eviQ Cancer Treatments Online. http://eviq.org.au [cited 2013 Jul 10]