Saving Drug Development Time with Digital Pathology

Why do I get my clinical data in weeks, and my pathology data in months? The local tissue response is a key diagnostic component, containing valuable efficacy and toxicity signals only decipherable by expert, credentialed pathologists. But getting information from these tissues is tedious and time-consuming. Clinical chemistry is automated, flow cytometry is high-throughput, even high content screening of cells has been nearly completely automated. But when it is time to do tissue work, whether angiogenesis studies in discovery on xenografts, or rat liver toxicity GLP studies, or HER2 protein expression in Phase III oncology clinical trials, months slip by. The reason for this? Good tissue based research requires experts trained in tissue based research - pathologists. There are never enough pathology experts to go around, and they are rarely near where the tissue is. These pathologists make their observations with a microscope, then record them manually for others.

Companies who understand and value their pathologists and tissue trained biologists have substantial competitive advantages in the pharmaceutical industry.  There have been trends recently to "automate pathology", or "automate tissue biomarker measurements" or other approaches designed to leave these experts out.  Anyone who works in solid tumor research knows that it isn't just that a biomarker is expressed, but where it is expressed that matters.  Which tissue, which pattern, what is the pathology diagnosis?  At Aperio we assume that the pathologist, whether a veterinary pathologist in preclinical toxicity studies, or a clinical anatomic pathologist in clinical trials, will remain central to the road to new oncology drugs for many years to come.  Our goal is to make pathologists more efficient.

Aperio is a rapidly growing technology company that has created and leads the market segment of digital pathology.  We produce high resolution scanners, software and services that allow an entire glass microscope slide to be viewed on a computer screen from 20x to 100x oil resolution.  Of the 500 Aperio scanners sold in 38 countries, nearly 200 are in use to support the pharmaceutical drug development process -- at biotechs, pharmaceutical companies, preclinical and clinical CROs, and government institutions like the NIEHS, NCI, and NCTR. 

Eleven of the top thirteen largest pharmaceutical companies use Aperio's digital pathology solution in biomarker discovery and preclinical tissue-based research.  In outsourced toxicity studies, nearly all of the major preclinical CROs offering pathology services have adopted Aperio's digital pathology solutions. Slides are scanned at the histology lab location, and then viewed by one or more veterinary pathologists from any location.  Pathology working groups, informal peer reviews, archiving, and other essential collaborative activities can be conducted remotely with a full electronic record.  Because the entire tissue section is scanned as one large image, computer analysis can be run to quantitate toxicology lesion size or measure protein expression.

The critical driving factor to the widespread adoption of digital pathology in the pharmaceutical sector is time savings.  In this article we look at specific opportunities for time savings across discovery, preclinical, and oncology clinical trials, and encourage executive decision markers to ask their vendors to demonstrate time savings as part of their technology value proposition. 

Time savings in Discovery
There are two goals in discovery with tissue-based research - get quantitative efficacy data on compounds, and get this as quickly as possible.  Cell-based screening leads tissue-based screens in throughput, even though tissue screening frequently provides more biologically relevant data about efficacy with cells in their native environment.  Any system that automates this process must build it upon pathologists or expert biologists, with the tissue and data flowing around their key observations.

A. Angiogenesis measurements in oncology biomarker research
Many anti-angiogenesis programs use microvessel density measurements in xenografts or matrigels for measuring efficacy.  This typically means weeks of manual counting by someone trained and supervised by a pathologist - few pharmaceutical companies in 2009 economics can afford to pay pathologists to do this manual counting.  The question is, can a computer be trained by a pathologist to more accurately and precisely count vessels than a non-pathology scientist?  In a recent presentation[i] by Novartis, Pathology Experts, and Aperio, two Ph.D. scientists manually counted vessels in matrigels and compared this with a whole slide scanning approach.  The scientists had 60% of the vessels in common, but had 40% where they disagreed.  Interestingly, each had 20% (one in 5) vessels which one scientist found and the other missed.  Overall they counted similar numbers, but they were not counting exactly the same vessels.  This approach was compared to whole slide scanning and automated whole section analysis.  The slides are scanned in 2 minutes at 20x and then only the vessels are found and counted with a microvessel analysis algorithm.  The computer by nature has perfect precision, and could be trained by the pathologist to be in agreement, with an error rate of 10%.  Other interesting morphology biomarkers, like cell wall thickness, lumen area, number of small vessels, vascular perimeter, etc. could also be calculated, something the scientists could not do.  The average time spent counting by the scientists was 10 times longer than the time required to scan and analyze by the computer, and the scanning and computer analysis can be fully automated and parallelized if required.

B. Amyloid plaque detection in neuroscience
New compounds in Alzheimer's are commonly screened against large numbers of rodent models to determine efficacy.[ii] A high-volume core imaging laboratory at Pfizer investigated the time savings with implementing whole slide imaging and analysis.  They reported that a 2.5 month rodent study analysis measuring amyloid plaque could be completed in only 1 month with Aperio's whole slide imaging solution[iii]. 

C. Cell counting in ophthalmology
Alcon Laboratories developed an all retina automated ganglion cell counting methodology utilizing Aperio's whole slide imaging and analysis.  These measurements are critical to glaucoma research, and require high volume application. The image of an entire retina was previously obtained by creating a montage of hundreds of photos, each captured individually. The whole procedure usually takes 5-6 hours but the newer method scanned the entire retina and counted retinal ganglion cell nuclei in 14 minutes.  With this method, the retinal ganglion cell count of an entire drug study can be completed overnight.[iv]

Time savings in preclinical - the right veterinary pathologist viewing the right slide
In preclinical research, there are substantial time savings associated with the adoption of digital pathology within a pharmaceutical organization, particularly when it is adopted across distant sites.  This allows pathologists on different continents but at the same pharmaceutical company to agree on scoring, discuss differences, and allow for more efficient distribution of pathology services.  

At Novartis Institutes for Biomedical Research (NIBR) in Cambridge, Massachusetts,  pathologists were challenged to provide timely analysis with quantifiable measurements for the many thousands of samples the company generates in both early research and later preclinical development phases. Pathologists analyzed biomarkers and microscopic tissue parameters to understand efficacies, toxicity and mechanism of action; these insights provided critical guidance to the drug discovery effort.  In their first two years at Novartis, Aperio ScanScopes scanned over 20,000 slides, nearly all of which were subjected to image analysis. Slides were scanned for a variety of uses, but the dominant use was high volume pre-screens of IHC-staining on paraffin embedded specimens.  The benefits of the Aperio systems were immediately recognized, as researchers saw an over 30-fold improvement in turnaround time.[v] The systems' automated nature and entire-slide approach made for a streamlined workflow that enabled researchers to focus on analyzing the results, rather than on taking many tedious steps to obtain results. Faster turnaround time ties directly to research efficiency and costs, making it easy for Novartis to see an immediate return on its ScanScope system investment.  Also beneficial to researchers was the fact that algorithms provided a means to quantitate effects and eliminate subjective estimates or counts to better the drug research development process. Results could be provided with more confidence in the underlying methodology used to achieve them, and could be more readily compared across different projects, for additional research impact.  The use of digital pathology has now expanded to three continents and four locations, in the United States, Europe and China Novartis operations.  Novartis can standardize on terminology across a global organization, and is seeing tremendous benefits from early adoption of the technology.[vi]

A second top 10 pharmaceutical company generates 10,000 slides per year in its preclinical operations.  The company focused on substantially improving its efficiency, quality, and throughput with the adoption of digital pathology from Aperio.  Approximately 6 months after installing the last of the initial three ScanScope systems, the organization has seen a 25% reduction in turnaround time for standard tissue slides, and a 50% reduction for tissue microarray slides.  By using digital slide conferencing for its peer reviews, the organization estimates it is saving $10,000 per review.[vii] Pathologists are also pleased to have reduced work impact, since they no longer have to travel and lose what can be days of work for one meeting.  

At contract research organizations, the time savings in preclinical GLP studies are also substantial.  At the beginning of a study, the veterinary pathologists and toxicologists can agree on scoring systems and toxicology nomenclature without having to travel or mail slides.  During the study, adverse effects can be quickly discussed with digital slide conferences.

Digital Pathology Improves the Efficiency and Accuracy of International Oncology Clinical Trials
Oncology clinical trials are expanding internationally while laboratory tests become increasingly complex.  Pathology is a key diagnostic component of clinical trials used to measure efficacy signals that are decipherable by expert, credentialed pathologists.  As more trial sites are added and fewer patients are enrolled at each site, global pathology test standardization is critical. Digital pathology facilitates international clinical trials by allowing the pathologist instant access to histopathology data globally, as well as quantitative image analysis tools to standardize test results. Pathologists working remotely can support international enrollment and the expansion in test complexity, making their test results more quantitative and reproducible. 

In response to the clinical needs in drug development, Aperio has recently launched a new product, Digital IHC (immunohistochemistry) for Oncology Clinical Trials. Digital IHC is being offered by select industry-leading contract research organizations, as it leverages their existing large immunohistochemistry test menus. Slides are scanned at remote clinical sites, and then uploaded to a secure HIPAA and 21 CFR 11 compliant database.  Pathologists can conduct confirmatory diagnosis during enrollment, discuss results with distant colleagues or study directors, or complete full electronic pathology workflow from slide to case report.  Aperio's recently lauched Genie^TM histology pattern recognition technology can be used to measure protein expression specifically in the cancerous cells of interest, automatically ignoring adjacent stromal tissue. Even with patients from hundreds of locations and dozens of international sites, the pathology assessment is centralized and standardized across all study participants. 

Instant global secure access to pathology data will contribute to faster oncology clinical trials. Logistics will improve, with the ability for an anatomic pathologist at a pharmaceutical company to do confirmatory reading during patient enrollment at a remote international location without the costs of time and travel.  Communication improves between the pathologists at the CRO and the study directors and oncologists at the pharmaceutical sponsor. Searchable whole-slide pathology databases post-study will become a standard tool in the near future, similar to how virtual compound databases are used in computational chemistry today. Quantitative results are derived from the entire tissue section, not small "representative" areas photographed using a microscope. For example, Targeted Molecular Diagnostics, a CRO recently acquired by Quintiles, uses Aperio's Digital IHC to multiplex downstream phosphomarker measurements as efficacy readouts in kinase inhibitor clinical trials.[viii]

The local tissue response to treatment is a key diagnostic component at every stage of drug development, containing valuable efficacy and toxicity signals decipherable by expert, credentialed pathologists. Adoption of digital pathology results in remarkable time savings across discovery, preclinical, and clinical trials. Pharmaceutical organizations should expect substantial returns on investment in faster times from this technology, and should measure these time savings as the technology is implemented.

Footnotes:
[i] Potts, SJ et al. Performance of a novel automated microvessel analysis algorithm across whole slide digital images. Society of Toxicologic Pathology Annual Meeting 2008.
[ii] Lillard-Wetherell, K. 2008. Whole slide analysis in Alzheimer's research: application note.  Available at www.aperio.com.
[iii] Milici, AJ et al.  Comparison of Plaque Burden in Tg2576 Mice Using the Aperio and TurboScan Imaging Platforms. Pfizer Global Research & Development. Pathology Visions Conference 2008.
[iv] Yaacobi, Y et al. A Novel, Automated, High Throughput, Retinal Ganglion Cell Counting System. Alcon Research. ARVO 2007.
[v] Novartis NIBR Application Note. Available at www.aperio.com.
[vi] Deeds, J, and H. Gardner. Multi-Site Integration with Digital Pathology: Three Countries, Four Time Zones, One Oncology. Pathology Visions Conference 2008.
[vii] Aperio: Top 10 pharma technical note. Available from www.aperio.com.
[viii] Hill, J. The Use of Image Analysis in Assessing Biomarkers: Implications to Clinical Trials, Drug Development and Patient Treatment. Targeted Molecular Diagnostics. Pathology Visions Conference 2008.