The Pharmaceutical Pathologist

Big pharma may not seem like a natural home for the anatomic pathologist; historically, our function has been to diagnose disease through examining human or animal tissues, but our expertise in the mechanisms of cellular and tissue response to injury means we’re well placed to play an important role in delivering drugs to the clinic. Drug development – from the “bench to bedside” – needs an ever-increasing number of disciplines and technologies to combine to bring a molecule to market.

To launch a new pharmaceutical, regulatory approval must first be secured – key to this is demonstrating therapeutic efficacy (how well does the drug work compared with others?) and safety (what are the potential adverse effects?). Toxicologic pathologists are crucial to this process, working on the front line to address both of these questions during preclinical studies. If you’re unfamiliar with drug discovery, this may seem like a foreign landscape, but it utilizes all of the same tools and skills pathologists routinely apply in clinical practice and research – namely, the assessment of gross and microscopic changes combined with technical aids like electron microscopy, immunocytochemistry, in situ hybridization and image analysis.

The role of toxicologic pathology in the pharmaceutical industry is perfectly demonstrated by the development of several new classes of oncologic drugs that target tumor-derived blood vessels (angiogenesis) and tumor proliferation – two of the six hallmarks of cancer (1). In each case, careful “down the microscope” observations contributed to the proof-of-principle, thereby increasing confidence that they could translate into a meaningful medicine that might go on to affect the lives of thousands of patients.

Proving theories behind therapies

Tumor reliance on angiogenesis may seem obvious to us today, but it was not until 1971 that the hypothesis was proposed by the scientist Judah Folkman. Writing in the New England Journal of Medicine, he suggested that tumor angiogenesis presented a new therapeutic approach for fighting cancer (2). Over thirty years later, this idea was validated by the approval of several new molecules, including bevacizumab, sunitinib and sorafenib. So what role did the pathologist play in these landmark approvals?

Using a combination of CD31 immunostaining of vascular endothelium and image analysis, toxicologic pathologists proved (at least in human xenograft tumors) that treatment with these compounds resulted in reduced vascular density and size, and therefore inhibited tumor growth.

As anti-angiogenics were being developed, the research community was also exploring other areas of tumor biology, searching for the all-important “druggable target”. The most notable of these was the epidermal growth factor receptor (EGFR), a tyrosine kinase receptor that, when activated (i.e. mutated), provided a growth advantage to non-small cell lung cancer cells. Treatment with EGFR inhibitors resulted in reduced xenograft tumor growth in mice, and histological analysis using EGFR phospho-specific antibodies demonstrated that this was associated with dephosphorylation of the receptor – i.e., inhibited activation. However, the first-generation of anti-EGFRs weren’t perfect, because adaptive tumor resistance almost always developed: commonly a T790M point mutation that renders the receptor insensitive to inhibition. Fast forward a decade, and third-generation anti-EGFRs were being created – these irreversibly bind to T790M, solving the problem. To test this, standard xenograft models were supplemented with more sophisticated transgenic mouse models which inducibly expressed the mutated T790M receptor in type II pneumocytes.  These mice reliably developed multiple lung adenocarcinomas (3). Regression of established tumors only occurred with third-generation inhibitors – and once again, histological imaging was able to demonstrate that reduced tumor mass was associated with dephosphorylation of the mutant EGFR, and inhibition of the downstream signal transduction cascade.

Finally, pathologists have also played a key role in the development of a newer class of agent: the aurora kinase inhibitors. Aurora kinases are known to play an important role in chromosome alignment and segregation, and cytokinesis. They are over-expressed in a number of human malignancies so targeting them should (theoretically) result in the tumors failing to correctly divide, leading to apoptosis. This should be observable because tumor cells would be expected to become tetrapoid or even octapoid because of disruption of the mitotic machinery. Clearly, histological assessment was ideally placed to provide physical evidence of targeted molecular inhibition. When the experiments were performed, the correlation with the theory was uncanny. Significant effects on tumor growth were observed, along with degenerating cells that showed unmistakable signs of nuclear enlargement – proving that the drugs were working as expected.

Ensuring safety and predicting problems

When developing therapeutic molecules, efficacy is nothing without safety, and toxicologic pathologists are ideally placed to inform safety assessments. Like all pathologists, we’re trained to assess the response of tissues to all types of noxious stimuli, including drug-induced toxicity. During the development of vascular endothelial growth factor receptor (VEGFR) inhibitors, the toxicity profile was first established in rodents, where pathologists noted a triad of changes characterized by ovarian atrophy, growth plate dysplasia and incisor tooth dental dysplasia (4). The affected organs are all dependent on a continuously growing vascular supply, and once the process was inhibited, degeneration or atrophy of the organ was an inevitable tissue response. In addition, some other changes were noted, two of which – hypertension and increased glomerular mesangial matrix with proteinuria – had important implications for humans. Hypertension was associated with improved clinical outcome, whereas proteinuria was associated with poorer survival (5).

Similarly, when EGFR inhibitors were being developed, preclinical assessment identified epithelial atrophy as one of the major toxicities associated with this class of compound; a direct result of receptor inhibition in epidermal tissues. In rodents, this change was characterized by thinning of the skin and an associated follicular dysplasia, inflammation and redness. In the clinic, patients treated with these drugs often develop epidermal rash, and this is also thought to be associated with improved prognosis and response (6).

Preclinical histological assessment isn’t only for characterizing adverse effects – it can also predict them before irreversible damage occurs. Take the example of FGFR/MAPK (fibroblast growth factor receptor/MAP kinase). During development, widespread metastatic mineralization was a significant preclinical concern. In order to choose potential treatments with a low risk of toxicity, the presence of mineralization could be correlated with changes in serum biomarkers (such as calcium or phosphate concentration), and then used in a predictive way to select new molecules, or even to screen patients for early signs of mineralization.

The need for safety, as well as the development of patient-specific efficacy biomarkers, is sure to continue to grow as personalized medicine becomes mainstream in our hospitals. It’s likely that toxicologic pathology will play a big part in the early development of many of them.

An evolving role

It’s inevitable that, going forward, existing technologies like image analysis and in situ slide techniques will become more sophisticated. New technologies (such as microRNAs) and genome editing techniques (like CRISPR-Cas) are going to be introduced into the drug development process – and toxicologic pathologists, like those in any field, will have to adapt and specialize in order to keep pace. Despite this, I think pathology as we know it, is here to stay: tissue responses to injury do not change, and our work provides a window into the physical reality of how drugs affect cells. However, our role and our purview are likely to evolve as we use new and creative ways to validate in vitro techniques, help develop biomarkers, and aid in integrating and interpreting the clinical relevance of bioanalytical data. The roadmap may be changing, but the role of the toxicologic pathologist is likely to remain front and center in the drug discovery process.

Peter Hall is a toxicologic pathologist at AstraZeneca, Alderley Park, Macclesfield, UK.