So just how did James Bond, Moby Dick and the London Natural History Museum teach me about oncology in clinical trials and tumor behavior? Here’s my story. As a child I loved wandering through the London Natural History Museum (NHM). It was fascinating to experience the natural world through its exhibits which was made all the more exciting, knowing these corridors had housed the research and development department of covert operations of the SOE (Special Operations Executive) during the Second World War. Legend has it that the real life director of the department was the inspiration for the character “Q” in Ian Fleming’s James Bond series.

The NHM has now added a modern five story annex named the Darwin Center which is dedicated to the life sciences and works to enhance the public understanding of science.  In the mid 1970’s, the exhibits on Evolution and Darwin’s work were relegated to one of the darker and more remote galleries of this Gothic Victorian building.

Principles of Tumor Behavior and Biology

Understanding evolution had become an obsession in the manner of Captain Ahab’s quest for the great white whale. Thankfully the cause was not a sadistic biology teacher but the inability to grasp the concept that explains the panorama of life.  I think my mother, a biologist, despaired of me! I’m glad I persevered because much of tumor behavior and biology can be explained via its principles(1). For example, drug resistance can be explained on the basis of three principles:(2-4)

• Individual variation exists in the population
• Variations are heritable
• Variations in individuals lead to differential survival and reproduction (2)

In other words, it is possible to select out resistance in tumor cell populations by the very act of treatment. By chance a random clonal cell line will develop or inherit a genotype that allows it to survive in the harsh environment of a tumor under treatment. Resistance mechanisms don’t possess intelligence but respond through pathway modulation that is rarely predictable. In fact, the presence of several clonal proliferations mediated by evolutionary mechanisms has been demonstrated in individual tumors (4).

Imaging in Tumor Behavior

Biopsy sampling becomes a real problem in the presence of spatially distinct colonies. Imaging of course has the distinct advantage of being non-invasive and probing the live organism rather than sampled tissue. Limitations of course exist. Clinical modalities cannot resolve microscopic processes. This is the realm of molecular imaging which promises increased resolution and sensitivity beyond the length scales of current scanning equipment.

In 1859 Darwin proposed an interpretation of life based on natural selection. So far, no scientific data have reliably refuted observation. Rather it continues to eloquently explain the complex behavior of cancer. My thanks to Ian Fleming, Herman Melville and a dark forgotten corridor in the NHM that has me entertained for 30 years!

Sources:

1. Greaves M. Darwinian medicine: a case for cancer. Nat Rev Cancer 2007;7(3):213-221.
2. Gerlinger M, Swanton C. How Darwinian models inform therapeutic failure initiated by clonal heterogeneity in cancer medicine. Br J Cancer 2010;103(8):1139-1143.
3. Casanovas O. Cancer: Limitations of therapies exposed. Nature 2012;484(7392):44-46.
4. Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366(10):883-892.

Remember when investigative sites would write down on three-part paper the information we needed to know, randomize patients via envelopes, and have investigator meetings in exotic locations?  Clinical data managers were responsible for making sure the data was entered into the database correctly, and checking the quality of the data.  Those were certainly the good ‘ole days.  And they are most certainly gone.

After attending the recent Society for Clinical Data Management conference, it is obvious the role of the data manager has greatly expanded over the years. They now need to be clinical data management technology experts so they can debug integration issues, fix edit check problems, unravel randomization matters, and perform site support.  Clinical data managers also need to be experts in database design, current CDISC standards, all regulatory guidances, and vendor relationship management. Throw on Six Sigma expertise and study metrics available, and it’s easy to see why data managers could be a little overwhelmed.

“How many of you want to answer queries in bed with your iPad?” This question from Nidhi Saxena during her presentation CDM: 10 New Technologies that will Change the way we Manage Data was not taken seriously, but it should be.  There are iPad apps to manage queries, randomize subjects, and find eligible patients.  The clinical data management technology is pretty amazing, if we can all keep up.

Clinical Data Management Technologies and Vendors at SCDM 2012

If you are shopping for systems, the vendor area can certainly help.  There were 21 vendors with EDC products in the vendor area ready to show off their systems.  The BioClinica “Randomize and Win a Prize” promotion was a big hit with three big gift card winners and lots of “I’m ready to try a different angle” t-shirts. The Data Driven awards highlighted the vendor’s latest clinical data management technologies: web services, integration, workflows, mobile platforms, automation, cloud-based technology, visualization tools, eSource collection via tablets, hubs for images, documents, and data could all be found as entries.

Since when did clinical data managers become responsible for the study monitoring?  Risk- based monitoring is a hot topic with an entire section devoted to case studies and approaches.  I really liked how Thomas Grundstrom of Quintiles focused on the “risk” part by talking about the 29 areas of risk that they have identified to determine which sites need more monitoring and who can get away with the basics. 

Real time alerts, true integration between systems, voice enabled interfaces, smart iPhone apps, personal health records, RFID tracking.  Are we over-engineering our clinical trials?  Is all this clinical data management technology really helping us to gain efficiency and quality or introducing more problems?

Stay tuned for a summary of the FDA Panel discussion from SCDM 2012 coming soon…

Technology? Data management? Free t-shirts? What's not to like about SCDM 2012! I'm looking forward to attending the Society for Clinical Data Managers conference September 22-24 in Los Angeles. Below is a summary of trends we will be hearing about along with some commentary by yours truly.

The SCDM 2012 conference will feature two and a half solid days of presentations. There is a huge focus on clinical trial data management technology and how it is changing the role of data managers and how to take advantage of this technology to run studies more efficiently. Several years ago, many data managers were worried that the use of EDC would eliminate the need for a data manager or combine the roles of data management and monitoring. The reality is no one's role has been eliminated. These advances have streamlined the data management process to create faster lock times and reduce quality issues. That's a big improvement!

Adaptive Clinical Trials

There is a track on Adaptive Trials, which we are seeing more of, but the definition has changed a bit from what was originally considered “adaptive.” When the concept was first introduced, we envisioned trials where we could see the data in real time and make decisions that changed the scope of the study by adding and removing efficacy parameters or changing the treatment schedule as we learned what was safe and tolerable. These days, most of the adaptive trials we see have to do with changing the dosing options in a very controlled manner. We don't see a lot of study design changes based on adaptive decisions and we rarely see anything adaptive outside of a Phase 1 or Phase 2 study. There is a lot more potential here, but sponsors find it a challenge to outline the possible options in the framework of a protocol. Perhaps we can introduce a rolling protocol that would offer a lot more adaptability? Imagine the reaction to that proposal!

Endpoint Adjudication

The track on Endpoint Adjudication is a very exciting use of clinical trial data management technology we are seeing these days. Using permissions and workflow control, it is simple to present the correct data to an adjudicator to collect their responses. Using alerts and bringing data from multiple sources, we can easily automate the adjudication process within the EDC system without having to use a special technology.

eSource and Risk Based Monitoring

We continue to hear talk of using eSource to eliminate the source verification role, and we are also seeing presentations on risk based monitoring. We have customers that do both of these already and the data management technology appears to be the easy part. The challenging part is coming up with internal processes where the sponsor is willing to accept the risk and process changes. The risk to using eSource is minimal since entering data directly into the EDC system without transcribing it to a source doc first improves quality. The challenge here is letting go of that almighty paper source doc. Believe it or not, there are people out there that don't want to trust all their data to technology and need to have that piece of paper. Risk based monitoring involves the risk of not comparing every single CRF to its paper based source. The risk here is minimized by randomly selecting what gets checked and providing corrective action if a site has too many errors. That said, most of the risk based monitoring still has the monitor review 100% of the critical variables within a study. In our conservative industry, this is as risky as we get!

I can't end this blog without putting in a plug for BioClinica's role in the SCDM 2012 conference. Jonathan Andrus, BioClinica's Senior Vice President of Operations, will chair a session focused on collaboration between data management and clinical monitoring professionals to improve data quality and increase the speed and quality of study management.

In addition to meeting me, visitors to BioClinica's booth #123 can enter our "Randomize and Win a Prize" Trident IWR/IVR demonstration.  Everyone who enters will either win a T-Shirt, a 4GB Flash Drive, or a $100 Gift Card. Hope to see you at SCDM 2012!

As a follow-up to my prior blogs about clinical apps for use in drug studies, I would like to direct readers to a patient education series produced by Orca Health which I find particularly useful, at least in my therapeutic area.

Of particular note are the spine and knee clinical apps that provide good training on basic anatomy. Users of these apps can fully explore the spine and knees by visualizing the anatomy in 3D and rotating it in space. Furthermore, these clinical apps allow the user to articulate in the knee or spine through the full range of motion when viewing in any aspect.

The spine app goes beyond basic movement and allows users to take two vertebrae and expand them out to see the bony anatomy, intervertebral disc and nerves and see the interplay and positioning. Both clinical apps also have image sets, mainly of MRI to show the anatomy as radiologists observe them.

Each clinical app features a tab that lists the most common medical conditions associated with the body part and labeled MRI scans showing the issue and the disease state. With the knee, the apps have embedded movies showing the arthroscopic surgical procedures of the various ligament and cartilage repairs.

There are now a number of other medical and clinical apps available that provide insight into the various organs and body systems. My wife, who is a pediatric nurse and a clinical tutor, downloaded a number of apps to improve the teaching experience on the ward. The one on the kidney was particularly clever from my perspective as it allowed you to zoom into the micro anatomy.

While not free, these apps are a very good value at around $5 each and are available through the iTunes store under the name “ORCA MD.” There is neotenous change about to occur in teaching medicine, anatomy, and physiology. Students will learn by seeing and experiencing the body through interactive clinical apps as opposed to traditional textbook study. I will certainly be using these apps to teach our non-medical teams the basics of anatomy so they can more fully understand the software they are developing. By putting the pieces into context, these elements become more than just words on a test script or software development line.

Please let me know if you have any other clinical apps to recommend that could be helpful or stand out to you.

One of the driving forces in medical imaging is the innovation that is taking place in medical imaging software.  I previously wrote about this from the 2010 Radiological Society of North America (RSNA) meeting.

As an example, medical imaging software can now accurately align PET and CT or even MR images from any major manufacturer and fuse them together, even if they have been acquired at a different time or place.  Combined PET/CT scanners have been around for a few years, but Siemens recently launched the first PET/MRI scanner, which really required some hurdles to be overcome and combined with the possibility of fusing the images, provides some interesting improvements in diagnostic capabilities.

Tri-modal fusion of an MRI scan with a PET/CT scan, or fusion of standalone PET to MRI offers non-expert medical imaging centers the potential to generate diagnostic insights that can help with oncology tumor assessment, staging and delineation.

Innovations in Medical Imaging from Mirada Medical Software

A company at the forefront of this is Mirada Medical Software.  BioClinica recently adopted their software to enable us to offer even more powerful and robust medical imaging analysis tools for quantification and tracking of findings in clinical trials.  MD Anderson Cancer Center announced late last year that they had adopted the software, so we are in good company.

When we compared competing medical imaging software products, what made the Mirada software particularly appealing to me was:

• Its robust history of development
• Ease of integration with a database
• Application to multiple therapeutic areas such as oncology, neurology, cardiovascular
• Ability to undertake review with multiple time points showing  

Mirada’s latest XD3 software release includes volumetric CT segmentation in addition to tools for CT/PET and MR/PET to facilitate both anatomical and functional response tracking.

Picture Credit: Mirada Medical Software

Some of the advanced features include:

• Support for multi-timepoint gated, multi-sequence MR, and multi-phase CT data
• Software based PET/MR
• Support for unlimited number of simultaneous follow-up scan review
• Support for standard disease response quantification workflows such as RECIST and PERCIST
• Built in support for disease specific workflows such as Head and Neck and Melanoma
• Data QC for automatically warning for inter-scan variations
• Voxel intensity classification tool for identifying tissue based on density within a region
• Casebook for presenting volumetric DICOM datasets in Microsoft PowerPoint presentations

You can find out more information about their products by visiting the Mirada Medical  website.

The Mirada software is already being integrated into our central lab workflow.  I’m excited about the additional functionality and capabilities this will offer clients. If you’d like any more information about these innovations in medical imaging software, please do not hesitate to contact me.

Dr. Miller will attend the 19th International Bone Densitometry Workshop August 20– 24, 2012 in Breckenridge, Colorado. BioClinica is proud to sponsor this event.