Lab Soft News

Radiofrequency Ablation of Lung Cancer Found Effective

I came across a recent article that discusses a non-surgical means for ablating malignant lesions of the lung for patients who cannot tolerate surgery or other therapeutic approaches (see: Non-Surgical Approach Can Treat Lung Cancers). It's called radiofrequency ablation (RFA), Below is an excerpt from the article with boldface emphasis mine:

A minimally invasive procedure normally used to treat liver cancer also holds promise for lung cancer patients, according to a new study. In the study..., 88 percent of lung cancer patients responded well to treatment with percutaneous image-guided radiofrequency ablation (RFA). RFA is performed in less than an hour and is a non-surgical procedure that targets large tumors with no harm to surrounding healthy tissue. After one year, 70 percent of patients survived at least one year with few side effects; none that impaired lung function, the researchers report....Lung cancer is the leading cause of cancer death in both men and women. Surgery is the standard treatment for early-stage, non-small-cell lung cancer (NSCLC), which constitutes about 80 percent of most malignant lung tumors. Unfortunately not all patients are eligible due to other health reasons. The alternatives, radiotherapy or chemotherapy, do not have good survival rates. The study...involved 106 patients with malignant lung tumors that were smaller than 5 cm in diameter. Thirty-three patients had NSCLC; 53, metastatic lung cancer from the colon; and 20, metastatic lung cancer from other sites in the body. All the patients had been turned down for surgery, radiotherapy or chemotherapy. The major post-RFA complications were pneumothorax (27 instances) and pleural effusion (4 instances) which needed drainage.

Here is a more detailed description of RFA from RadiologyInfo:

In radiofrequency ablation, imaging techniques such as ultrasound and computed tomography (CT) are used to help guide a needle electrode into a cancerous tumor. High-frequency electrical currents are then passed through the electrode, creating heat that destroys the abnormal cells.

Basically, this technique is analogous to "cooking" the primary or metastatic lung lesions in a microwave oven. This approach would seem to bypass both the thoracic surgeons and pathologists unless fine needle aspiration (FNA) is used to diagnose the NSCLC lung lesions. In the case of those patients with metastatic disease, biopsies may have been obtained previously to diagnose the primary lesions. This seems to be an excellent approach for patients who are not candidates for other therapies. It also effectively demonstrates the expanding scope of practice of interventional radiology.

BioImagene Launches a Contract Research Organization (CRO)

I have posted a number of previous notes about various contract research organizations (CROs), particularly Covance that has special expertise in supporting the clinical lab testing associated with clinical trials. I have also posted previous notes about the growing popularity of holding clinical trials offshore in countries such as India (see: Status and Challenges of Offshore Clinical Trials; When Clinical Trials Go Awry). I was interested to learn recently that BioImagene, a provider of digital pathology and life sciences services has launched a CRO business (see: BioImagene Launches AgilityBio, a New Generation Contract Research Organization Leveraging Its Innovative Digital Pathology System). Below is an excerpt from the article with boldface emphasis mine:

BioImagene...launched AgilityBio -- an integrated contract research organization (CRO). AgilityBio offers preclinical and clinical services in partnership with well established Indian CROs. It provides its clients with a US-based customer liaison team to facilitate management of international projects....AgilityBio’s integrated service offerings include preclinical services, Phase I - IV clinical and diagnostic trials and biomarker discovery for the pharmaceutical and biotech industries. It has the advantage of using BioImagene's web-based digital pathology technology, to provide access to image data anywhere, any time. 

This initiative strikes me as interesting. This new U.S.-based CRO entity, AgilityBio, partners with established Indian CROs and brings special imaging expertise to the game. So, in essence, CROs seem to be getting more complex and forming networks with each of the network partners contributing a specific set of skills. I must say, however, that I was a little mystified by the assertion of the BioImagene web site that "In the life sciences industry, two-thirds of the data is in the form of images." This strikes me a being a little high but I suspect that they are including document images in their definition of the total number of images.

Some Interesting Insights into Companion Diagnostics

I have posted a number of previous notes about companion diagnostics (see, as one example, the following: Companion Diagnostics Gaining Ground, But Slowly). Now comes an interesting article that describes in detail about the partnership between a diagnostics company and a pharmaceutical in the future development of such a companion lab test. Below is an excerpt from the article with boldface emphasis mine:

The area of companion diagnostics - ideally defined as a diagnostic developed in tandem with a drug to screen patients for clinical studies and then be commercialized alongside the drug for diagnostic and treatment purposes - has been getting more attention since last year's approval of Selzentry (maraviroc) from New York-based Pfizer. That drug, designed specifically to target HIV patients who test positive for the CCR5 receptor, was developed with the help of Trofile, an HIV co-receptor tropism assay, from Monogram Biosciences. The companies began working together in 2002, and Pfizer used the assay to determine which patients would most likely to respond to treatment for enrollment in clinical studies. In 2006, South San Francisco-based Monogram licensed to Pfizer global rights to Trofile. In that case, "Pfizer came to Monogram," [an analyst] said. "They were developing an entry inhibitor for CCR5 and wanted to select patients appropriately for the trial." And that's the way it likely will have to work between pharma and diagnostics firms in the future, with pharma responsible for driving the development. "For diagnostics firms, it's difficult for them to develop a test without seeing the drug on the other side," [he] said. "It's tough for them to fund that innovation. You wouldn't have seen Monogram developing Trofile on its own, hoping that somebody comes out with a CCR5 therapy."

Here's another quote from the same article:

But [the analyst also] said one of the biggest hurdles is convincing pharmaceutical firms to jump on board the diagnostic train, mostly because pharma has been reluctant to move away from their blockbuster-drug approaches, selling a single pill that can reach a broad market. "Drug companies can be somewhat cautious that having a companion diagnostic might limit the drug's label," .... So "pharma and diagnostics firms aren't really on the same page yet."

All of this makes perfect sense. Big Pharma has been cautious about the wide availability of companion diagnostics in the past because of the belief that such lab testing may constrain the sale of a particular drug. However, as the goal of developing blockbuster drugs begins to fade, Big Pharma now finds it useful to approach diagnostics companies to develop a companion test in parallel with the drug development and clinical trials in order to select the most appropriate subjects for clinical trials and also patients downstream. As I composed this note, I could not help thinking about my previous note (see: Moving Resources from the Therapeutic to the Diagnostic Silo) about using lab tests to monitor therapeutic efficacy. Companion diagnostics provide one of the means by which the right drug can be selected for the right patient.

"Eminence-Based" Surgical Pathology and the Digital Pathology Department

In a recent note, I commented on the new strategic alliance between GE Medical and the University of Pittsburgh Medical Center (UPMC) in the pursuit of the digital pathology department and whole slide imaging (see: GE Medical Partners with UPMC in Pathology Imaging Venture). It is often stated that digital radiology took about a decade to mature and that digital pathology will take an equal amount of time to become the accepted standard of practice. However, major incentives were available in the conversion to digital radiology such as the ability to offer new imaging procedures with attractive profit margins plus a groundswell of enthusiasm on the part of hospital clinicians for these new offerings. These same incentives do not exist for digital pathology -- there are no additional profit margins to be gained for the hospital and the pathology reports to clinicians are largely the same except for the routine integration of digital images into surgical pathology reports.

However, it you are searching for a "killer app" associated with digital pathology, it is most certainly image search. By this I mean the ability to isolate "fields of interest" in a challenging surgical pathology case, an unusual malignant tumor for instance, and then match them against a large image database of diagnosed lesions for similar lesions. For challenging cases today, this same process often takes place laboriously by searching surgical pathology atlases on the shelf. Parallel to this process, many such cases are also referred to local colleagues for their opinions and also sent to "marquee" surgical pathologists who have established reputations as having superb diagnostic skills in various specialty areas. This is the basis for what has been called eminence-based surgical pathology.

And now comes the rub. These marquee pathologists have little to gain and much to loose by the introduction of digital pathology and image searches. Image searches and pattern matching is, in fact, what is taking place in the brains of these eminent pathologists. I suspect that they will have little enthusiasm for any technology that serves to lessen their influence, prestige, and livelihood. Many of these latter pathologists are also highly placed in the hierarchy of prestigious pathology departments. I suspect that they will be leaders of the chorus opposing the conversion to digital pathology and perhaps highlighting the failures and inadequacies of digital pathology departments.

GE Medical Partners with UPMC in Pathology Imaging Venture

GE Medical and the University of Pittsburgh Medical Center (UPMC) are launching a pathology imaging business (see: GE, Pittsburgh hospital establish imaging business). Below is an excerpt from the article:

General Electric Co....is investing $20 million in technology that will allow doctors to share and transmit images of microscopic human tissue....GE Healthcare and the University of Pittsburgh Medical Center are each putting up $20 million to establish Omnyx LLC, a business to develop and commercialize technology allowing doctors to store and display on computers, digital images of human tissue from microscope slides, allowing colleagues anywhere to participate in consultations. Jeffrey Romoff, president and chief executive of the University of Pittsburgh Medical Center said the deal "puts together the people — GE — that know the technology and the market with us, who know the science and the patient care."... "Pathology imaging is a natural extension that GE would likely want to get into," said Gene Cartwright, chief executive of Omnyx. ...Omnyx promises to come up with a product in about two years that will speed up scanning materials from a slide into a digital file from between two and five minutes now to about half a minute, he said.

I have published a number of previous notes about the entrepreneurship of UPMC. Dr. Mike Becich and his colleagues have also established the pathology department at UPMC as a national leader in not only pathology imaging but all of pathology informatics. From my perspective, the deployment of practical whole slide imaging systems is a key to the future success of surgical pathology.

Research Study Documents that Obesity Cause Heart Inflammation

I have posted a number of previous posts about the risks of obesity with emphasis on diabetes and the metabolic syndrome (see: Metabolic Syndrome: The Problem You May Never Have Heard About). A recent article focusing on the relationship between obesity and inflammation of the heart raised issues that were new to me (see: Obesity-Related Inflammation Boosts Heart Risks). Below is an excerpt from it (boldface emphasis mine):

Obesity causes prolonged inflammation of heart tissue that in turn boosts heart failure risk, according to a U.S. study of almost 7,000 people. The latest findings from the Multiethnic Study of Atherosclerosis (MESA) are believed to provide the first large scale of evidence of such a link and give the estimated 72 million obese American adults another reason to change their lifestyle....There is "now even more reason for (obese people) to lose weight, increase their physical activity and improve their eating habits," [said an author of the study]. He and his colleagues tracked the development of heart failure in an ethnically diverse group of nearly 7,000 people, ages 45 to 84, who enrolled in the MESA study, starting in 2000. Of the 79 participants who've developed congestive heart failure so far, 35 (44 percent) were physically obese (body mass index of 30 or greater).[The authors] also found a link between inflammation and metabolic syndrome, which doubles a person's chances of developing heart failure. On average, obese participants were found to have higher blood levels of key immune system proteins involved in inflammation (interleukin 6, C-reactive protein, and fibrinogen) than non-obese participants. A near doubling of average interleukin 6 levels alone was associated with an 84 percent increased risk of heart failure....Metabolic syndrome is a collection of risk factors -- obesity, high blood pressure, elevated blood glucose levels, excess abdominal fat, and abnormal cholesterol levels -- that increase the risk of heart disease and diabetes.

In a previous note, I discussed how bariatric surgery has now been shown to "cure" diabetes (see: "Curing" Diabetes with Bariatric Surgery). The now proven link between obesity and heart inflammation suggests that such surgery may also prevent, or delay, the onset of congestive heart failure. I had always assumed that heart failure in the obese was related to an overworked organ rather than to intrinsic heart problems such as inflammation. I suspect that elevated inflammatory biomarkers in obese patients will increasingly be used as a clinical indication for bariatric surgery, particularly using an adjustable gastric band which can be inserted by laparoscopic surgery and therefore results in less operative mortality and morbidity. The device is now recommended for patients with a body mass index (BMI) between 35-39 and co-morbid conditions such as diabetes, hypertension, and hypercholesterolemia.

Integration of Anatomic and Clinical Pathology

Although I have posted a number of previous notes about the potential merger of pathology and lab medicine with radiology, I strongly believe that such a change must be preceded by a much tighter integration of clinical pathology (CP) and anatomic pathology (AP). Part of the value of the proposed merger for radiologists will be the ready availability of a total view of disease based on both molecular diagnostics and morphologic observations coming from the pathologists.

As I have noted before, one of the key practice models for this future direction for CP and AP will be the practice of hematopathology in which both the morphologic characteristics of malignant cells as well as their biochemical nature are taken into consideration when arriving at a diagnosis (see: Reinventing Pathology: The Hematopathologist as a Model for the Pathologist of the Future). In addition, hematopatholgists frequently participate in the selection of therapy for patients because such choices are frequently based on their diagnoses and thought processes.

There is growing evidence that the practice of surgical pathology is now moving closer to clinical pathology. Evidence for this can be found in the lectures of Dr. Jeff Myers. He has emphasized the close collaboration of surgical pathologists with pathology informaticians to increase patient safety (see: Aligning Surgical Pathology & Aligning Surgical Pathology & Informatics to Promote Informatics to Promote Patient Safety). Research in tissue biomarkers will also help to convert surgical pathology to a more quantitative discipline (see: In-Vitro Biomarkers vs. In-Situ Biomarkers; Changing Strategies for Interrogating Tissue Samples: A Systems Pathology Primer).

In my past blog notes and in the interest of being all inclusive, I have found myself using the awkward phrase pathology and laboratory medicine to refer to the field. This is truly a mouthful but I can't come up with the better name for the more closely merged CP-AP unit that I am discussing here. Therefore, I have decided to refer to it in the future as PLM. I know that change is merely cosmetic but having a more manageable name will be useful. If and when PLM merges with radiology, this problem will go away. We can then refer to the merged entity as diagnostic medicine.

Service-Oriented Architecture in Healthcare: The End of Hierarchy

I have posted previous notes about the federated model in healthcare computing. Drs. Balis and Routbort recently lectured on the relevance of this model, closely linked to service-oriented architecture (SOA), at Lab InfoTech Summit 2008 (see: The Value of a Federated Architecture in Pathology: Test Order Entry;The Value of a Federated Architecture in Pathology: Test Result Reporting). For me, the most appealing aspect of this approach to healthcare computing is that it eliminates hierarchy among systems. Under the federated model in a hospital, each information system becomes a single-source-of-truth (SST). I recently encountered an article in the Financial Times in which an advocate of SOA made some very powerful arguments in favor of this approach (see: What IT means to me: Software will bring about the end of hierarchy). Below is an excerpt from it with boldface emphasis mine:

In [the view of Hubertus “Hub” Vandervoort], technological developments march in lock-step with social change. He foresees a world where new data processing methods – essentially service-oriented architecture (SOA) – will force the abandonment of old, hierarchical management practices and the adoption of ways of working based on trust and commitment....SOA has been hailed as the technology that will save the data centre and redeem corporate computing, but the fact is that outside the cognoscenti, it is one of the most obscure of the IT industry’s welter of three-letter abbreviations...[He describes it in the following way:] “The key things are that it is loosely coupled, contract-oriented interfaces. That is what it is from a technology or computer science standpoint. It is sometimes better described through what it enables. It enables, in theory, complete heterogeneity – total isolation of implementation. I don’t have to think at all about what you choose to implement your technology on and if I conform to the standards available in SOA, we should still be able fully to interoperate.....SOA is really the first time we can create true multi-party interactions. Client/server technology, mainframe technology even web technology was essentially point-to-point. The web gave us the ability to connect to millions or zillions of points but only one at a time. SOA enables us to create composite applications that can be built up from the technology provided by 20 parties. From a user experience standpoint, I think I’m dealing with one application but in fact I’m blending in real time the interactions of potentially dozens of third parties.”

For me, the federated model/SOA is the only way to go. Hospital CIOs and LIS directors are being buried by the need to create custom HL7 interfaces among often dozens of heterogeneous specialized systems, each of which may be "best of breed" and required by physicians working in their specialty areas. Think instead of the inexpensive and quickly deployed "loosely coupled, contract-oriented interfaces" described above.

It would be unlikely that a CIO managing a hospital EMR would be willing or able to refuse to develop an SOA contract with the hospital LIS director at a time when some 70%-80% of hospital diagnoses are based on lab information. Concentrate your thoughts on the desirability of the end of hierarchy in healthcare computing and you will be ahead of the game.

Status and Challenges of Offshore Clinical Trials

I have posted a number of previous notes about the globalization of clinical trials and the contract research organizations (CROs) that provide lab support for such trials such as Covance and Charles River. Clinical trials are an important market for the clinical lab industry. A recent article about the globalization of clinical trials (see: MIT Study Quantifies Globalization Trends) provides some additional insights into this trend. I provide an excerpt from it below with boldface emphasis mine:

Outsourcing of biopharmaceutical clinical trials to China and India is growing at a substantial rate, but in real terms the much-ballyhooed nations are still "very minor players," [according to an MIT professor]. [The U.S.] commands a 48.7 percent share of total trial activity and has eight times the number of trial sites of second-place Germany....India, a growing global hub for trial-related technology, is also well positioned to become a major clinical trials player....Trial density, the proportion of recruiting sites relative to overall population, is greatest in the U.S., Canada, and several Western European countries....But it's becoming substantial in some Eastern European countries such as the Czech Republic, Hungary, and Estonia. Presumably, this makes the region increasingly able to offer a competitive number of sites suitable for global trials....Pooling data from ethnically and culturally diverse populations may become problematic with the march toward personalized medicine and pharmacogenomics...and at some point may even reverse the current globalization trend. Drug-naïve patients on vegetarian diets may also be differentially affected by classes of drugs commonly used in the Western world. Further, the integrity of the informed consent process may be jeopardized in nations where the physician-patient relationship is more "hierarchical."

Despite the possible cost advantages of conducting clinical trials offshore in less developed countries, I agree with the general drift of this article that the majority of them will continue to be conducted in the U.S. One reason is the challenge, as noted above, of conducting clinical trials in countries like India where the genetic character of the research subjects may differ from the majority of patients in the U.S. There are other scientific and ethical barriers to overcome with offshore clinical trials as reflected in the following 2004 article (see: Indian Guinea Pigs for Sale: Outsourcing Clinical Trials):

"When getting a subject's informed consent, some research is complex and it is difficult to convey the relevant issues," notes US-based bioethicist Ruth Macklin who has participated in the development of various international ethical guidelines for collaborative research in developing countries. Equally worrisome is the fact that "people may not distinguish between treatment and research. There is a false belief that sometimes research may have a direct benefit. Research does not provide individualized medical treatment, titrating doses according to the patient's need, for example."

As a former member of a hospital institutional review board (IRB), I found the review of the occasional offshore studies that came to the committee to be very challenging. The motivation of impoverished people to enroll in clinical research studies can be far different than subjects in developed countries like the U.S.

Fantastic 1819 Anatomic Illustrations by Kyoto Physician Yasukazu Minagaki

The Tohoku University Library in Japan provides an online display of the painfully-real Kaibo Zonshinzu anatomy scrolls created in 1819 by Kyoto-area physician Yasukazu Minagaki. The style of these medical drawings is markedly different than the more sanitized approach favored in Western countries (see: Kaibo Zonshinzu Anatomy Scrolls Online).

More details about the these Japanese anatomic drawings are supplied at the Pink Tentacle (see: Kaibo Zonshinzu anatomy scrolls (1819):

Unlike European anatomical drawings of the time, which tended to depict the corpse as a living thing devoid of pain (and often in some sort of Greek pose), these realistic illustrations show blood and other fluids leaking from subjects with ghastly facial expressions. The fact that the bodies used in scientific autopsies in Edo-period Japan generally belonged to heinous criminals executed by decapitation adds to the grisly nature of the illustrations.

These works of medicine and art from 1819 are worth a look and somewhat shocking even for physicians who may have been trained using Gray's Anatomy, now available on-line. The online display of the drawings requires you to click the "left arrow" to navigate through the scroll.