The cancer therapy market is evolving dynamically. We understand more and more about cancer biology and we are now able to develop drugs that selectively target the molecular drivers of tumorgenesis. Joanna Przewrocka examines a development in the field – polypharmacology – in relation to the paradigm of ‘one companion diagnostic – one drug’.
All this sounds great—approaching cancer from multiple angles using cocktails consisting of many targeted agents in order to suppress it more effectively—but is this at all feasible within the current cost structures? Targeted therapies are priced at premium rates similar to those charged for therapies previously designated with orphan or small market status. If we use a cocktail of targeted agents do we also combine the prices of all the drugs used? Will payers be able to cope with the costs?
Although researchers are eager to test the targeted therapy combinations, they are confronted with many barriers. One is commercial competition. Pharmaceutical companies worry that side effects that emerge during combination therapy may influence the commercialization of a drug that would be otherwise safe when used alone, especially if the agents are still under development and have not yet been approved. It is difficult to predict the efficacy and toxicity of rationally designed drug cocktails in pre-clinical models, even when the individual agents have already shown clinical anticancer activity. For example, pre-clinical experiments supported the combination of gefitinib and trastuzumab in breast cancer, erlotinib and bevacizumab in renal cell carcinoma, and cetuximab and bevacizumab in colorectal cancer. However, efficacy of all of these combinations failed expectations in clinical trials.
Cocktails consisting of many targeted agents, particularly ones that target different pathways used by normal cells, can also induce unacceptable side effects. Lung cancer researcher Jeffrey Engelman thinks that for this reason, patients will be able to tolerate a cocktail only for a short period of time. He suggests putting patients on a single drug, then intermittently giving them a ‘pulse’ of a cocktail for several days.
What about the diagnostics? The ‘one companion diagnostic – one drug’ paradigm may not work out so well due to a couple of reasons.
First of all, cancer is extremely heterogenic. This can be observed not only between different types of cancer, but also between tumors of two different patients with the same type of cancer, between two metastatic lesions within one patient, and even between two different cells within the same tumor. Although some tumor types are associated predominantly with abnormal activity of a single kinase (such as in chronic myeloid leukaemia, where 95 per cent of patients have a chromosome abnormality known as Philadelphia chromosome, that results in the translation of bcr-abl fusion proteins), it is not as obvious in other cancer types. For example, a study by Sequist et al. (2011) scanned genetic profiles of tumors from 552 non-small cell lung cancer patients and identified the presence of different mutations (KRAS, EGFR, translocations involving ALK, BRAF, TP53, PIK3CA, CTTNB1, NRAS, HER2, IDH1), sometimes occurring simultaneously with others. Therefore,the entire cancer genome should be scanned to identify all known mutation and tailor molecularly guided therapies.
Secondly, genomes most probably would have to be scanned multiple times to monitor disease progression, as cancer cells are constantly evolving and developing resistance to anti-cancer drugs. A 2011 study published in Science Translational Medicine (STM) by Engelman’s group on non-small cell lung cancer, illustrates this complexity by analyzing 37 biopsy samples from patients with the EGFR mutation who were given Iressa or Tarceva and later became resistant. Although some resistance mutations were known, others were new, and for 30 per cent of the samples his team could not identify the mechanism. Some tumors even morphed into a different type of lung cancer that requires an entirely different treatment. In addition, biopsies from three patients collected during the course of treatment showed that some tumors that developed resistance mutations later lost them. Engelman’s STM study suggests that clinicians would need to constantly biopsy patients and devise one complex cocktail after another, each tailored to a patient’s evolving tumors.
Next-generation sequencing (NGS) technologies (as opposed to first-generation dideoxy ‘Sanger’) have the ability to massively parallelize the sequencing of millions of DNA templates and are now considered by many laboratories for routine diagnostic use. Compared with other sequencing methods NGS has improved sensitivity, speed and cost. There are currently  three companies offering NGS platforms: Roche, Illumina and Life Technologies. While those commercial platforms are readily available, the analysis of the NGS data is not a simple task and could potentially become a challenge when implementing NGS into standard clinical practice. There are several commercial analysis software programs, some of which are provided by the platform providers. There are multiple steps involved in transforming raw sequence reads into a final list of clinically useful variants. Several different algorithms have to be used in order to improve upon base calling accuracy and reduce systematic errors, read alignment (e.g., BWA, AQ, Bowtie and Novoalign algorithms), and separate real variations from sequencing noise (e.g., Bayesian algorithms).
All in all, the list of novel cancer biomarkers will hopefully be expanding and cancer research will lead to new solutions and strategies. Pharmaceutical companies will soon need to accommodate for the increasing speed of cancer research, which might involve abandoning the concepts of high price tags for targeted therapies and the ‘one drug – one companion diagnostic’ paradigm which don’t appear to be sustainable in the long run.