Lab Soft News

Progress in Diagnoses with Endo-Microscopy

Image from Wikipedia

I have a special interest in emerging technologies that may ultimately compete with histopathology in diagnosing pathologic lesions from tissue samples. I have posted a number of notes about molecular imaging but never any about endo-microscopy. A recent article about this topic caught my attention (see: Endo-microscopy Technique Shows Promise for Early Colon CA Diagnosis) and below is an excerpt from it, with boldface emphasis mine. Shown on the right is an image of a flexible endoscope.

...we [previously} profiled an endo-microscopy system from Paris-based Mauna Kea Technologies...[T]his company's main product, the Cellvizio fibered confocal microscopy system, allows a "practitioner to insert one of the miniprobes (only 300 um to 2.8 mm in diameter) into a conventional endoscope and record microscopic level movies of the tissue as fast as 12 frames/sec." The latest news is that Stanford University scientists tested the system in a small study, to evaluate the detection of early stages of colon cancer, and they were quite pleased with the results....The [scientists] found a short protein that sticks to colon cells in the early stages of cancer. Before screening a person, they spray that short protein attached to a fluorescent beacon into the colon. The protein then gloms on to any cancerous cells and creates an easily visible fluorescent patch. They then used a miniaturized microscope called Cellvizio GI, developed by Paris-based Mauna Kea Technologies..., to peer inside the colon and look for those telltale spots. Not only did the researchers see fluorescent patches, they could make out the individual cancerous cells.....In the initial trial with 15 patients, the technique detected 82 percent of the polyps that were considered cancerous by a pathologist.

Endomicroscopy, alternatively referred to as in-vivo microscopy, is defined in the following way: a newly developed endoscopic modality, which allows in vivo microscopy of the mucosal layer in about 1000-times magnification with subcellular resolution during ongoing gastrointestinal endoscopy. As noted above, the technique can be supplemented by a ligand protein + fluorescent label to enhance the ability to detect malignant transformation of colonic polyps real-time during GI endoscopy.

It's interesting that researchers are now combining the techniques of endomicroscopy with a fluorescent-labeled ligand to identify malignant transformation of polyps. In essence, the technique is being transformed into a type of molecular imaging managed by a gastroenterologist rather than a radiologist. However, in this former case, there is a human observer of the gross lesion supplemented by microscopic observation using a built-in confocal microscope. From the patient's perspective, the advantages of in-vivo real-time microscopy is that the malignant lesion(s) can be immediately detected and resected.

CLMA Annual Meeting in Atlanta Canceled Due to Tornado Damage

I may be the last person in the country to have learned this, but the CLMA annual meeting scheduled for Atlanta on March 29-April 1 has been canceled. Here's the announcement from the CLMA's web site:

CLMA has just been notified by officials at the Georgia World Congress Center that the site of our conference will not be available due to serious safety issues caused by damage resulting from the March 14 tornado that struck downtown Atlanta. Unfortunately, CLMA has been forced to cancel ThinkLab `08, scheduled for March 29 through April 1, 2008.

The Georgia World Congress Center is putting on a brave face, according to the following excerpt from an article (see: Tornado damage won't cancel multi-day events):

At the Georgia World Congress Center, the scars are still visible from the March 14 tornado that ripped through Atlanta: Missing ceiling tiles, plywood covering empty window frames and gaping holes in the roof. The storm left eight of the GWCC's 12 exhibit halls unavailable. Still, GWCC officials are confident that, except for possibly one convention starting Sunday, no multi-day events will have to take their show — and their money — elsewhere....Only the Clinical Laboratory Management Association, scheduled to start [march 29] and bring in about 6,000 attendees for two nights, has been canceled.

I am sure that this problem has caused serious challenges for the CLMA, conference registrants, and exhibitors. I extend my best wishes to all of them.

Metabolomics, Metabolomic Profiling, and Systems Biology

It's hard to keep abreast of the "-omics revolution" but I have a sense that help is on the way. A recent article (see: "Metabolomics" Could Lead to Early Blood Test for Parkinson's) suggested to me that lab professionals will now be paying more attention to metabolomics. The article referred to metabolomic profiling for Parkinson's disease, a new term for me, and below is an excerpt from it (boldface emphasis mine):

A fairly new technique called "metabolomic profiling" may soon yield blood biomarkers for Parkinson's disease (PD), according to a report in the February issue of the journal Brain. Metabolomic profiling involves analysis of changes in thousands of molecules in body fluids and tissues....In ongoing research in PD patients and controls, [researchers] are using high performance liquid chromatography coupled with electrochemical array detection to study alterations in thousands of serum metabolites. "We discovered a clear differentiation between the metabolomic profiles of the Parkinson's disease patients versus those of the controls," said the lead author]. "No one molecule was definitive, but a pattern of about 160 compounds emerged that was highly specific to Parkinson's patients." While the significance of many individual metabolites to Parkinson's remains to be determined, changes in a few well-known metabolites linked to oxidative stress were clearly associated with Parkinson's, the team reports. These included significantly reduced concentrations of the antioxidant uric acid and an increase in another antioxidant, glutathione.

Metabolomics is the "systematic study of the unique chemical fingerprints that specific cellular processes leave behind" - specifically, the study of their small-molecule metabolite profiles. The metabolome represents the collection of all metabolites in a biological organism, which are the end products of its gene expression. Thus, while mRNA gene expression data and proteomic analyses do not tell the whole story of what might be happening in a cell, metabolic profiling can give an instantaneous snapshot of the physiology of that cell. One of the challenges of systems biology is to integrate proteomic, transcriptomic, and metabolomic information to give a more complete picture of living organisms.

Systems biology is defined as the field that seeks to study the relationships and interactions between various parts of a biological system (metabolic pathways, organelles, cells, and organisms) and to integrate this information to understand how biological systems function.

So, putting all of this together and drawing on the article cited above, the technique of metabolomic profiling will be used by researchers to establish a diagnostic picture (i.e., snapshot) of the molecular changes occurring in patients with diseases such as Parkinson's. Systems biology is the study of the relationship of all of the various components of any biological system and encompasses the metabolome. Abnormal systems biology would then involve the detection of abnormalities in patients with aberrant metabolomic profiles.

Dr. Jose Costa will be lecturing on strategies for interrogating tissue specimens for abnormal proteins at the Lab InfoTech Summit in Las Vegas on April 11, in essence an exploration of systems pathology. We have been quantifying serum proteins for many decades but are now entering an era when we can accurately characterize in-situ tissue proteins as well, particularly the abnormal ones.

A Closer Look at Companion Diagnostics Strategies

I have written a number of previous notes about companion diagnostics. Briefly stated, this is a strategy pursued by some IVD companies, Roche Diagnostics in particular, whereby the company develops a gatekeeper biomarker assay. This is a lab test that serves to qualify a patient for treatment with a particular drug. The most common example of such a test is the HER-2/neu assay that is required prior to treatment with Herceptin. Herceptin is used for adjuvant treatment of HER2-over-expressing node-negative (ER/PR-negative or with one high-risk feature) or node-positive breast cancer following multi-modality anthracycline-based chemotherapy. To restate the obvious, a companion diagnostics strategy can be very successful when linked to a popular drug but obviously less attractive if the associated drug proves not to be well accepted. Another interesting scenario occurs when a lab test can be used to provide evidence that a particular drug is not therapeutically effective -- essentially an anti-companion test.

Given this background, I pay attention when an executive of a national reference lab opines about the interplay between drugs and lab tests. Below is an excerpt from such an article (see: LabCorp CEO: Tests Could Slice Market for Cancer Drugs),an interview of David King of LabCorp (boldface emphasis mine):

Some drug companies are betting on the future of targeted medicine, hoping to improve outcomes by using high-tech tests to figure out which patients could benefit from a given drug. But what if a diagnostic test comes along and shows that a drug that’s already been approved for a disease may not work well for some patients? The pharmaceutical companies won’t always welcome this,” David King, CEO of the diagnostics shop LabCorp.....But with specialty drugs getting more expensive, and testing getting more sophisticated, such tests are likely to emerge soon whether drug makers like it or not. Genentech’s Avastin, which was recently approved for breast cancer without clear evidence that it extends the lives of patients with the disease, would be one potential candidate for such a test, King said. A new diagnostic test could wind up shrinking the market for a drug by ruling out some patients....Abbott [has announced that] its diagnostics unit has cut a deal with Genentech, OSI and Roche to develop a genetic test to assess the potential benefit of Tarceva, a drug for cancer of the lung and pancreas that is co-marketed by the companies.

I have the distinct impression that the term companion diagnostics may have outlived its usefulness -- but I am equally sure that it will not go away. My major problem with it is that the word companion sounds too warm and fuzzy. My first thought was that we should now refer to such testing as surveillance or sentinel lab testing. The basic idea here is that the lab test is used to determine efficacy or, contrariwise, the lack of efficacy or even the potential harmful side effects of treatment with the drug. I must also confess to a slight anxiety about too cozy a relationship between the developers of such surveillance/sentinel lab tests and the pharmaceutical companies that develop and manufacture a particular drug. Nevertheless and on balance, I think that such relationships will probably work for the benefit of the patients by determining whether they are candidates for newly developed drugs.

Old Blood Reported to Increase Heart Surgery Morbidity

I must confess that I had a visceral reaction to a recent article in the NEJM about the use of "old blood" in surgery (also see: 'Old' Blood Raises Heart Surgery Risk) because of my role as a blood banker in a previous life. All blood bankers dislike the term "old blood" because one of their primary roles is to effectively manage the blood bank inventory. This goal translates into reducing the number of units outdating on the shelf and this, in turn, generally mandates a FIFO (first in, first out) strategy rather than a LIFO (last in, first out) approach. Intuitively, however, I think that all parties concerned have understood that "fresh blood" is probably superior but that older blood for some patients was better than no blood.

One of my solutions to this conundrum was to provide "incentives" to surgeons to use less blood intraoperatively. I accomplished this goal in collaboration with Dr. Harold Oberman, the director of the blood bank at the University of Michigan Hospitals, with the development of the maximum surgical blood order schedule (MSBOS) whereby the maximum number of units crossmatched preoperatively and held in the OR refrigerators for various types of surgical procedures was capped. Many surgical procedures called for only a type-and-screen.

The underlying rationale behind this policy was that surgeons were less likely to transfuse blood in marginal cases if they had gone beyond the number of units on hand and needed to order more. I believe that the MSBOS programs that were adopted in multiple hospitals in the latter part of the1970s and 1980s provided an incentive for the movement toward "type-and-screen only" as the most common pre-op blood order for coronary-bypass-graft (CBG) surgical procedures. It's ironic that one of the strong advocates of this approach was Dr. Delos (Toby) Cosgrove, a cardiac surgeon from the Cleveland Clinic, and that the article cited above comes from the same institution.

Here is an excerpt from the article (boldface emphasis mine):

Heart surgery patients who get transfusions with blood stored for longer than two weeks have a higher risk of complications and death than patients who get newer blood, a new study finds. The risk of death following heart surgery was 30% higher among patients transfused with blood stored for longer than 14 days, researchers from the Cleveland Clinic reported. These patients were also more likely to suffer surgical complications, including longer airway intubation, kidney failure, and infection. The study...[from the] New England Journal of Medicine, is not the first to suggest a link between stored blood age and surgical outcomes....The FDA allows red blood cells to be stored for up to 42 days. The average time of storage is estimated to be 15 days in a 2005 government survey....About half of the patients had transfusions from blood that had been stored for 14 days or less, and half had transfusion from blood stored longer. The average length of storage was 11 days in the "new blood" group and 20 days in the "older blood" group....[The older blood] patients had higher rates of death prior to leaving the hospital following surgery (2.8% vs. 1.7%), intubation lasting more than 72 hours (9.7% vs. 5.6%), kidney failure (2.7% vs. 1.6%), and potentially life-threatening blood infections (4% vs. 2.8%).The death rate a year after surgery among patients infused with older blood was also significantly higher (11% vs. 7.4%).

I suspect that there is little likelihood of either (1) the national blood supply increasing rapidly, permitting ad lib use of "fresh blood" or (2) an effective all-purpose artificial blood substitute without any undesirable side effects coming to the market. Autotransfusion has also not proven to be a practical substitute for the use of banked blood. It seems, therefore, that the only course of action left for us is to closely examine the morbidity and mortality experienced by patients receiving older blood in studies such as the one above and determine whether there are ways to ameliorate these deleterious effects.

Cerner's Winona Health Project Featured on the PBS News Hour

I have posted notes in the past about Cerner's Winona Health Project (see: The Winona Project: Is This a RHIO Success Story?). The project received very laudatory coverage on the PBS New Hour last night (see: PBS News Hour to feature Winona's Health Information Technology). Cerner is certainly to be commended for having launched this community-wide demonstration project five year ago and sticking with it over the years to achieve a successful result. The town, hospital, and medical practitioners were hand-picked because of their willingness to serve as a guinea pig for this experiment, but this detracts little from the successful outcome that can be seen today. It's a pity that more communities have not emulated the electronic medical record success experienced in this small Minnesota town.

Four points from the News Hour piece jumped out at me as very interesting and worth repeating, so I now present them to you:

• The CEO of Winona Health, the town's 100-bed hospital, is quoted as saying that there were only three PCs in the entire facility five years ago when the project was launched. She cites this as evidence about how much progress in computer literacy has been achieved in a relatively short time.
• The statement is made during the program that much of the emphasis of the project since its inception has been on personal health records (PHRs). I believe that this is a bit of a misstatement. The concept of the PHR hardly existed when the project was initiated five years ago. My recollection is that there was the expectation that patients would have the opportunity to input some of their personal health information into the electronic medical record and review such records from home. What we now understand as web-based applications and the PHR did not exist when the project was launched.
• The point is made in passing, and worth emphasizing, that the Cerner system installed throughout the entire town lacks interoperability with the EMR system running at the neighboring Mayo Clinic. When residents of the town are admitted to that health system, their comprehensive electronic records can't be viewed.
• Germane to this last point and also of great interest was the statement by one of the physicians that most patients in Winona have little interest in generating data for inclusion in their PHRs or even accessing their health records on-line. According to him, only about 10% of patients in Winona bother to use the system. Recall that this lack of interest occurs under very optimal conditions with access to a wide array of data. According to him, patients take a greater interest in accessing health information replicated to their PHRs from the hospital and physician office EMRs. This is sometime referred to as a "tethered" PHR system. I have made this same point in a number of my previous notes about PHRs.

Unelievable! -- A Replica of a Beating, Bleeding Heart

To "entertain" everyone at its 2008 Gala in New York, the American Heart Association displayed this artwork representing a beating, bleeding heart by Billy Chasen, an installation that's not meant for the squeamish but that all pathologists will certainly appreciate.

HPV Now Shown to Cause Oral Cancer in Men

I must admit that the research discovery described here now seems totally predictable -- it turns out that HPV is a major cause of oral cancer in men (see: HPV causing more oral cancer in men). In my mind, this predictability became obvious once HPV was shown to be the major cause of cervical cancer since the oral mucosa is also susceptible to infection by the virus and is similar to the stratified squamous epithelium of the cervix. Below is an excerpt from the article (boldface emphasis mine):

The HPV virus now causes as many cancers of the upper throat as tobacco and alcohol, probably due both to an increase in oral sex and the decline in smoking....The only available vaccine against HPV, made by Merck & Co. Inc., is currently given only to girls and young women. But Merck plans this year to ask government permission to offer the shot to boys. Experts say a primary reason for male vaccinations would be to prevent men from spreading the virus and help reduce the nearly 12,000 cases of cervical cancer diagnosed in U.S. women each year. But the new study should add to the argument that there may be a direct benefit for men, too....Previous research by Gillison and others established HPV as a primary cause of the estimated 5,600 cancers that occur each year in the tonsils, lower tongue and upper throat. It's also been known that the virus' role in such cancers has been rising. The new study looked at more than 30 years of National Cancer Institute data on oral cancers. Researchers categorized about 46,000 cases...[and] concluded the incidence rates for HPV-related oral cancers rose steadily in men from 1973 to 2004, becoming about as common as those from tobacco and alcohol. The good news is that survival rates for the cancer are also increasing. That's because tumors caused by HPV respond better to chemotherapy and radiation, Gillison said.

Here's an additional quote about the topic of oral sex and its association with HPV infection that was published in Time Magazine (see: Oral Sex Can Add to HPV Cancer Risk):

Of the 300 study participants [in a recently published study], those infected with HPV were also 32 times more likely to develop [cancer of the tonsils or at the base of the tongue] than those who did not have the virus. These findings dwarf the increased risk of developing this so-called oropharyngeal cancer associated with the two major risk factors: smoking (3 times greater) or drinking (2.5 times greater). HPV infection drives cancerous growth, as it is widely understood to do in the cervix. But unlike cervical cancer, this type of oral cancer is more prevalent in men.

Am I the only one who is disturbed by the fact that the number of entertaining but safe diversions is shrinking faster than the global icecaps?

Additional Discussion About Reinventing the Autopsy

In a recent post (see: Reinventing the Autopsy: CT Imaging as a Routine Part of the Procedure), I suggested that this may be the right time to begin to reinvent the autopsy. In particular, I believe that a whole-body CT scan should be a mandatory first step in all such procedures. At the beginning of this note, I cited Dr. Jared Schwartz as the individual who had stimulated my interest in integrating imaging techniques such as the CT scan into the standard autopsy. He has posted a comment to this note which I copy below with boldface emphasis mine:

Bruce you are correct I have been preaching to pathologists, pathology educators and hospital leaders of the potential value of reinvigorating the autopsy using modern technologic tools. Imaging combined with fine needle aspiration and the promise of molecular diagnostics could result in a minimally invasive autopsy or no autopsy in the traditional sense in some cases. Lets not forget other tools such as laparoscopy. High resolution CT imaging for autopsy is occurring at a few research/military facilities in a few countries. Unfortunately i have not been able yet to convince our health system to be an alpha site for this procedure. We need to continue to try and educate healthcare leaders that a modern autopsy using all the tools available would without doubt renew the autopsy as an important part of improving the quality of health care. BUT unless a financial model can be developed to show the savings through improved outcomes few hospitals will invest in the research needed to demonstrate whether our belief of imaging improving the autopsy and autopsy rates is correct. I am an optimist and thus convinced that soon a few brave institutions will step forward.

Jared is absolutely correct that a new financial model is required if the catopsy, the reinvented autopsy that will include medical imaging, has any chance of being widely adopted. Let me put forward a couple of thoughts with regard to the development of this new financial model:

• I stated in my original note that tissues harvested at the time of catopsy would be submitted, whenever possible, to tissue biorepositories and that the revenue from these biorepositories would be used to offset some of the costs of the catopsy service. I do not know whether this financial model is feasible or how much additional revenue would be raised by pursuing this option.
• I believe that majority of catopsies would require little additional work beyond the initial performance of the CT scan and its rapid interpretation. Given that such a scan would cost what I would estimate to be one-quarter of the cost of a standard autopsy today, I believe that converting an autopsy to a catopsy service would actually result in a cost savings. However, catopsies may be in such demand that their numbers would increase. This increased workload would obviously increase the total cost of the proposed catopsy service.

Finally, I want to speak to Jared's last point which relates to the need to demonstrate to hospital executives that the conversion to a catopsy service (or even the current autopsy service, for that matter) will improve health outcomes in hospitals. My working hypothesis is that a catopsy will yield more useful feedback for physicians about their care of patients than the current autopsy. I further hypothesize that these physicians, in turn, will improve their modes of practice (i.e., improve health outcomes) on the basis of the reports that they receive following the procedure. In order to acquire data documenting any improved outcomes as a result of the conversion to a catopsy service, I would suggest adoption of the following procedure:

After receiving the catopsy report, the responsible physician will be asked, as part of the hospital QC program, to comment whether he/she will modify future treatment of patients in any way on the basis of information contained in the report. If the catopsy report documents findings that differ significantly from those recorded in the ante-mortem medical record, such a response will be considered mandatory.

Because these responses from physicians will be viewed as an important component of the total hospital QC program, they will be treated in a confidential and protected manner. They will thus be identical to other queries that are used to improve the quality of care and identify possible judgmental or procedural errors by physicians or nurses.

   

Quintiles Promotes Its "Globally Harmonized" Lab Network

I have devoted a number of previous notes to contract research organizations (CROs) such as Covance and Charles River. In prior notes, I have discussed how Covance has developed a global lab network (see:  A Look at the World's Largest Central Laboratory Network). Quintiles now is emulating Covance in this regard but with a slight twist by promoting itself as "the largest wholly owned CAP-accredited laboratory network in the world" (see: Quintiles Central Laboratory in India Receives CAP Certification). Below is an excerpt from the article (boldface emphasis mine):

Quintiles Transnational Corp. today announced that its central laboratory in India has been certified by the College of American Pathologists (CAP). The central lab in Mumbai began supporting clinical trials in the first quarter of 2007...."With the certification of the India laboratory, Quintiles Central Laboratory Services now has the largest wholly owned CAP-accredited laboratory network in the world. This tightly controlled network with CAP- certified laboratories in the US, Europe, South Africa, China, India and Singapore allows us to support global and regional trials in almost every country in the world, [said a Quintiles executive."... In addition to laboratory testing, the Mumbai lab provides project management and logistical support, study specific laboratory kits for sites, local investigator support and samples management, storage and processing using consistent global methods. All laboratory data collected is globally harmonized and included in the Quintiles Laboratory Information Management System (QLIMS) and the protocol-specific database.

I suspect that the U.S. reference labs most likely to operate globally in a manner similar to CROs are those involved in genomic/proteomic testing such as Myriad. However, this segment of the market differs markedly from the CROs. To support medical research in India, Quintiles has set up a central lab in Mumbai where routine tests for regional clients can be performed quickly and inexpensively. Nevertheless, Quintiles and other CROs also need to enable the performance of the same tests in other labs around the world but still within its own lab network. This requires "harmonization" (or normalization) of testing whereby a particular test result can be interpreted using the same reference range regardless of where the test is performed in the network.

With regard to molecular genomic/proteomic testing such as that performed by Myriad that I have discussed in previous notes, I believe that there is little likelihood that such tests will be performed, at least in the short term, in any lab other than the the company's central lab in Salt Lake City. Quality control and performance issue are so critical for such tests that it does not make sense to attempt to operate out of multiple labs. The company will simply scale up the test volume in their home lab as test demand increases. For such esoteric testing, test turn-around-time is also not a critical factor so that a day or two of shipping time is easily tolerated with the final reports quickly returned to clients electronically.