Lab Soft News

Correlation Between Tumor Size and Serum Biomarker Levels

I believe that at some time in the very near future, we will be able to diagnose a wide variety of tumors with great specificity on the basis of their serum proteomic signatures in the serum. I must admit, however, that I have paid insufficient attention to the quantitative assessment of serum biomarkers as opposed to their qualitative value. A recent paper (see: Cancer Screening: A Mathematical Model Relating Secreted Blood Biomarker Levels to Tumor Sizes) therefore caught my attention and provided some some early hints about how the titer of a  circulating biomarker can be used to provide additional clinical information:

This study introduced a linear one-compartment mathematical model that allows estimation of minimal detectable tumor sizes based on blood tumor biomarker assays. Assuming physiological data on CA125 and PSA from the literature, the model predicted detection limits of tumors that were in qualitative agreement with the actual clinical performance of both biomarkers. The model may be helpful in future estimation of minimal detectable tumor sizes for novel proteomic biomarker assays if sufficient physiologic data for the biomarker are available. The model may address the potential and limitations of tumor biomarkers, help prioritize biomarkers, and guide investments into early cancer detection research efforts.

Two conditions were assumed in the study:

(1) the compartment (plasma) is well-mixed and kinetically homogenous; (2) the tumor biomarker consists of a protein that is secreted by tumor cells into the extracellular fluid compartment, and a certain percentage of the secreted protein enters the intravascular space at a continuous rate.

 I find the idea of estimating the minimal detectable size of a tumor based on biomarker titers to be intriguing. My vision of integrated diagnostics in the future is that multiplexed biomarker panels will be used for cancer screening of the general population or to assess symptomatic patients. Positive results from these panels can then be used to select the most appropriate medical imaging studies to confirm the diagnosis. This imaging process will surely be made easier if some idea of the minimal anticipated size of the putative tumor in addition to the organ site can be communicated to the radiologists interpreting the studies.

Cognitive Surplus and Its Relevance to the Clinical Lab Industry

Clay Shirky, mentioned in a previous blog note (see: A New Model for Medical Conferences: Broad Choices and No Waiting), has hatched a new concept about the allocation of brain power by individuals. He calls it the cognitive surplus (see: GIN, TELEVISION, AND COGNITIVE SURPLUS). Scroll halfway down the page at this link to watch the short entertaining video of his lecture on this topic.

Here's the basis of his idea in brief. He asserts that the industrial revolution forced a large portion of the population from the farms into cites, leaving many of them bewildered and disoriented. Initially, they turned to gin  to provide solace for their despair. Fast forward to the latter half of the 20th century with shorter work weeks and many people with time on their hands. The population then turned to television and sitcoms to entertain themselves and absorb their excess leisure time.

Because of the web, we have now entered what Shirky refers to as the "architecture of participation." Modern media, better stated as media on the web, has three drivers: consume, produce, and share. This is the reason why people have contributed what he estimates as 100 million person-hours, with no monetary compensation, to the development of the Wikipedia. This is time that participants have carved out of their cognitive surplus and allocated to this collaborative web-based encyclopedia. He also tells the story of the four-year-old child of one of his friends who was watching television with her dad. She then began searching behind the TV for something. Her father asked: "What are you looking for? She replied: "The mouse." Children now expect all media to be participatory.

So what is the relevance of all of this to the clinical lab industry? The lesson is simple and relevant to all sectors of the industry: IVD manufacturers, LIS vendors, and hospital/reference lab personnel alike. Let's say that an organization needs information from its customers or lab professionals at-large about the use of one of their products. Or let's say that the organization is designing a new product and needs input from the marketplace. What's the best way to obtain such input?

The answer to this question is to set up a lab-oriented web site with a participatory component and pose the key questions to participants. They will provide the necessary information if the web environment is entertaining, collaborative, and stimulating. They will provide these answers at no cost by "donating" a small portion of their individual cognitive surplus. Oh, one other thing. Corporate or lab executives who themselves do not spend a couple of hours per day on the web will have no idea how to create such an environment on the web. Moreover, they will probably squelch the ideas of anyone in their organization who actually knows how to do the job because they will be unable to identify such a solution as correct.

Vendor Credentialing and Compliance Web Site

Mr. HIStalk alerts us to a new vendor "purchasing, credentialing, and compliance" web site for hospitals:

Reader comment: "Re: Sisters of Mercy Health System, St. Louis. They are requiring vendors to register at vendormate.com and pay an annual fee to do business with them."

Mr. HIStalk Response: I hadn’t heard of Vendormate, which offers vendor credentialing and compliance solutions. It’s an ingenious business model run by mostly Georgia boys. Worth a look.

Here's how Vendormate summarizes its services on its web site:

Vendormate VISION provides a standardized process and directory through which organizations can better manage compliance and purchasing tasks associated with their suppliers. It will revolutionize how you qualify and choose your vendors and will be an essential daily tool across your organization, helping risk management, ethics, legal, contract management, executive management, and procurement.

(and)

Potential vendors answer general pre-screening questions and download and accept your policies. Existing vendors are asked additional questions based on your company's needs and are automatically sent through our patent-pending Vendor Quality Metric engine to be screened, credentialed, and scored based on criteria you select.

Sounds like a time-saver for hospitals and vendors alike. A very creative idea if the business model can be made to work. I suspect that the idea resonates more with hospitals than with their vendors even though it may save the vendors some elbow grease. My take on most vendors is that they would hate to screened by a "patent-pending Vendor Quality Metric engine" that would be out of their control (i.e. controlled by another vendor) and perhaps not cast them in the best light. If Vendormate is to be successful, it will be the result of many hospitals like the Sisters of Mercy Health System, referenced above, requiring participation by their vendors, or at least the ones over which they have some degree of leverage. Sole-source vendors can probably tell the good sisters to take a leap.

Wellness, Preventive Medicine, and the Classic Disease Model

When I was recently working on a lecture on integrated diagnostics and integrated diagnostic centers (IDCs) (see: The Evolution of Integrated Diagnostics into Integrated Diagnostic Centers), I became interested in the relationship between wellness, preventive medicine, and the classic disease model, the latter defined as the treatment of clinically overt acute and chronic disease. I subsequently presented this lecture, entitled Integrated Diagnostics Emerges as Key Element in Healthcare, with the references to preventive medicine appearing in slides 24 and 29.

Based on some of the ideas incubating in my mind during this process, I decided to develop a healthcare diagram with wellness located at the top and flowing down to chronic diagnosable disease at the bottom. This approach is somewhat analogous to the diagnostic workflow diagram that I developed for personalized medicine (see: Continuing Discussion of the Role of Lab Medicine in Personalized Medicine). The diagram that I came up with is displayed above.

Note that I have placed the Early Health Model (EHM) domain above the broken line with a relationship to the Wellness Domain that encompasses both Preventive Medicine and Genomic Medicine. I have done this because the EHM refers to the diagnosis of pre-clinical, pre-symptomatic disease and therefore lies outside of the classic disease model. Note also that Genomic Health frequently involves the diagnosis of a predisposition to disease and therefore also lies above the line.

My view is that there now exists an inexorable pull of patients, more accurately described as health care consumers, upstream and toward the Wellness Domain in the diagram. This is occurring partly because of the effects of science and technology on healthcare delivery (pre-symptomatic diagnosis) and also because of the striking interest of consumers in understanding and improving their present and future health status. Molecular diagnostics will play a major role to play in the development of this Wellness Domain because laboratory tests are not only used to evaluate the status of an existing disease but, when normal, also help to define wellness, which is to say the absence of diagnosable disease.

Lab InfoTech Summit 2009; March 16-18, 2009; Las Vegas

The first full-page advertisement for Lab InfoTech Summit 2009 will appear shortly in the September issue of CAP Today, including topical highlights of the workshop and the 15 lectures to be presented. Here's a PDF file of the ad if you want an early peak at it. One of the exciting new features of the conference will be LITS-Interop, a live demonstration in the exhibitor area of out-of-the-box connectivity across LISs from different vendors. Be sure to save the date for this major event.

Board Upheaval at the University of Maryland Medical System

I have posted a number of previous notes about how the non-profit status of some health systems is being called into question (see: Non-Profit Hospitals Drift from Their Mission Despite Subsidies), the unseemly profits being racked up by some health systems (see: UPMC Profits Up 77 Percent to $512 Million), as well as the "healthy" compensation of their CEOs (see: UPMC Rewards CEO's Entrepreneurship). Mr. HIStalk now reports about the recent political upheaval on the executive board of the University of Maryland Medical System. Below is his note with boldface emphasis mine:

The chairman and a third of the board members of University of Maryland Medical System resign as the organization struggles with governance between the medical school and the health system. This article says issues include doctor dissatisfaction with bottom-line emphasis and the governor’s appointing of board members without its input. It’s an interesting point: hospitals are one of few non-profits that operate under the business model, where they don’t pay taxes but have huge business-related income. Nearly all other non-profits are charities relying on outside support. When you think about a non-profit being a $2 billion dollar a year business like UMMS, that’s kind of weird, especially when hospitals that size sometimes pay CEOs $1 million or more a year in salary (according to tax records, the CIO job there pays $400K and UMMS, in fact, paid its CEO $2.6 million in the last tax year and some of the VPs are pulling down nearly a million).

The lede of a recent story in this Wall Street Journal article (see: Nonprofit Pittsburgh Hospital Starts a Company With GE) gets right to the point:

The University of Pittsburgh Medical Center is one of those business-savvy hospital systems that almost makes you forget the “non” in nonprofit hospital. Case in point: The health system is putting $20 million into a joint venture with General Electric to launch a new imaging company [see: GE Medical Partners with UPMC in Pathology Imaging Venture]....Like a multinational conglomerate, UPMC has health-care operations in Ireland, Italy and Qatar, we noted a few months back. Last year, UPMC spent more than $10 million on advertising, and paid its CEO more than $3 million. A spokeswoman told the WSJ the salary is in-line with other nonprofits of comparable scale and complexity.

So what's the answer? Perhaps I am being overly simplistic, but it seems to me that there should be some threshold percentage of revenue that nonprofit hospitals allocate for the care for the indigent and uninsured. This amount should calculated on the basis of the true cost of such care to the hospital and not the larger amount that they would ordinarily bill for the care. If this threshold is not reached, health systems such as UPMC should be taxed exactly like their peer "multinational conglomerates."

Update on 8/25/2008 at 6:06 p.m. ET

Check this out (see: UPMC profit down by more than $600 million):

Profits at the University of Pittsburgh Medical Center took a nose dive last year, falling to $5 million from $612 million a year before, officials reported in financial statements being released today. Executives at the region's largest health care provider attributed the 99 percent drop to a market-driven tumble in investment income and several hits on its operating income, including an $11 million payment to the Pittsburgh Promise scholarship fund. The 20-hospital network, which for the first time reached the $7 billion mark in revenue, remains on solid footing, officials said.

Lead Time Bias and Cancer Screening Tests

I recently came across an article (see: Early Test for Cancer Isn’t Always Best Course) whose title intrigued me. I wondered how in the world could early diagnosis of a malignant lesion result in a problem for a patient, setting aside the well-known issue of false positives. One of the major points raised in the article relating to lab screening was lead time bias. Here is an excerpt from it:

Another problem with determining the value of screening is that it results in “lead time bias.”....Any screening test can [also] lead to false positives, followed by invasive and risky tests. Large numbers of people often end up being poked, prodded and tested only to discover they’re fine.

This article was quite vague about lead time bias so I decided to seek a clearer definition for the term, which I quote below:

Lead time bias is the bias that occurs when two tests for a disease are compared, and one test (the new, experimental one) diagnoses the disease earlier, but there is no effect on the outcome of the disease-- it may appear that the test prolonged survival, when in fact it only resulted in earlier diagnosis when compared to traditional methods. It is an important factor when evaluating the effectiveness of a specific test.

Lead time bias thus only comes into play when one is comparing the results of two lab tests, a standard one and a new experimental one, perhaps a new multiplexed biomarker test for cancer. It seems to me that this is primarily a problem for researchers and biostatisticians. Ask 100 people on the street the following question: would you like to know that you are "cooking" a treatable malignant tumor earlier or later. I bet that 90+ would answer earlier.

The Mobile Web and the Future of eHealth

In a recent note, I commented on the fact that cell phones and particularly smart phones, have many of the features of computers and more attention should be paid to these devices for communicating health information to a broad swath of the population (see: Making e-Health Information Accessible with Smart Phones). In a recent story in the New York Times about Google's impending entry into the smart phone market (see: T-Mobile to Offer First Phone With Google Software), there was an interesting quote from Eric Schmidt, CEO of Google, about the mobile web:

Google thinks that many consumers will want to personalize their mobile phones with unique applications and services, including those made by Google. Google is eager to get the Android platform [the Google software to power mobiles phones] on phones quickly because it thinks that the mobile Web is vital to the long-term growth of its digital advertising business. “We can make more money on mobile than we do on the desktop, eventually,” Eric E. Schmidt, Google’s chief executive, said in an interview on CNBC this week.

Think for a minute what Schmidt is saying here. Google has built a business with a market cap of $158B based on access to the web using desktop and laptop computers. He believes that his company can make more money with the mobile web than the classic web accessed by these types of computers. Keeping this thought in mind and understanding that web sites oriented to health topics are some of the current favorites, how big will e-health topics be on the mobile web?

Continuing with this line of reasoning, it's now time, in my opinion, to begin to adapt consumer-oriented web sites about the significance and interpretation of lab tests to the mobile web. One good example of such a site is Lab Tests Online. Such modification would allow, for example, patients who have just completed a physician visit to use their cell phones to query this site and obtain a better understanding of their test results. I understand that many cell phone now have the ability to browse the web, but formatting problems for the smaller screens often make such browsing less than optimal.

Outsourcing of Clincal Trials of Drugs: Bringing Product to Market Faster

I have published a number of previous notes about outsourcing or offshoring of drug trials (see as one example: Status and Challenges of Offshore Clinical Trials). Conducting such trials in third-world countries can be both challenging as well as rewarding. A recent article (see: Should Clinical Trials Be Outsourced?) provides some interesting insights into this topic. Below is an excerpt from the article with boldface emphasis mine:

There are currently some 400 clinical trials underway in India, where the business is expected to be worth $1 billion to $1.5 billion by 2010. For western drug companies, it's a boon: India's vast pool of qualified, English-speaking doctors and lower labor costs make clinical trials up to 50% to 60% cheaper here. Beyond the savings in cash, India offers a crucial savings on time. A drug patent lasts 20 years; during the first seven or eight years of that period, typically, a new drug is tied up in several rounds of human clinical trials. That leaves little time for the drug maker to market and profit from the product. To speed up trials, pharmaceutical companies need to recruit large numbers of subjects quickly, something that is often not possible in western countries, where most patients have health insurance or government benefits and are reluctant to sign up. In India, by contrast, much of the population has no health insurance, and public health services are dismal; the promise of getting regular medical attention by enrolling in a trial is often too valuable to pass up. "Many patients in India enter into clinical trials because they have no better option to receiving care," says Sonia Shah, author of The Body Hunters: How the Drug Industry Tests Its Products on the World's Poorest Patients. "That is by definition a coercive situation."

I have discussed in the past the cost advantages of holding clinical trials in less developed countries such as India. I have also spoken about some of the difficulties in obtaining informed consent in such trials as well as the inherent coerciveness of offering healthcare to research subjects who aren't in a position to otherwise receive it or afford it. Recall that some subjects enrolled in double blind clinical studies will receive placebos so that treatment of their specific disease cannot always be guaranteed. However, I understand that it is also possible to compare a new treatment against a standard treatment in such a trial.

For me, the most interesting aspect of this article was that it highlighted one of the key advantages for pharmaceutical companies in holding clinical trials in third world countries. In such settings, it is far easier to recruit large numbers of treatment-naive subjects and the trial can thus be completed faster. Faster clinical trials potentially allow the company to bring a drug to market faster. The company therefore will then enjoy a longer period of protection under the U.S. patent laws if and when the clinical trial yields favorable results.

Differentiating Between RHIOs and HIEs

I have posted a number of previous notes about RHIOs. If you review these notes, you will discover that I have never been enthusiastic about RHIOs and their business models because I never thought that they offered sufficient incentives for hospitals to participate. John Moore, who blogs over at Chilmark Research, distinguishes between RHIOs and HIEs and predicts in a recent note, accurately in my opinion, why the latter organizations will have a greater chance of success (Why HIEs Succeed and RHIOs Languish). Below is an excerpt from his note with boldface emphasis mine:

Back at the beginning of the year, I did a Top Ten Predictions post where one of the predictions was the continued struggles of Regional Health Information Organizations (RHIOs) and the steady rise of Health Information Exchanges (HIEs).  Since much of the ballyhooed National Health Information Network (NHIN) is built upon the premise of RHIOs and their success, the NHIN is basically dead in the water. Sure, NHIN supported by RHIOs is a great and grand vision, but that is about it - a great and grand vision created by policy folks in government (and their academic counterparts) that has little basis in reality.  That divorce from reality stems from a lack of a true, justifiable business case.  There is simply no compelling business reason for the vast majority of health industry stakeholders to invest in RHIOs. (Note: those that have invested in RHIOs are most often doing it for political reasons, not business ones.) This is not the case for HIEs. These networks are created by entities, typically large hospitals, that do have a compelling business reason (one reason: encourages referrals to their hospital) to share health information within a network of business partners (their provider network).  This is quite similar to a supply chain network within the manufacturing sector....For hospitals its about delivering value to smaller practices, most often giving a physician a window into the HIE host’s EMR to facilitate care and insure referrals.

I agree with John's comments and conclusions. The primary goal of hospital executives is to increase their admission rates and they have very little interest in making clinical information more portable for patients. Although clinical belongs to the patients for whom it is generated, hospitals generally view it as proprietary to them. The consequences of the broad sharing of patient clinical information is to enable healthcare consumers to shop around among competing hospitals.

As John points out in his perceptive observations, the executives of a large hospital are very willing to reach out to their business partners such as those community physicians who refer patients to their facility. The assistance to such partners often takes the form of technical and financial support for physician office EMRs that are linked to the hospital EMR and can then provide continuity between the care delivered in physician offices with the care delivered during hospital admissions. The business case for such IT support is self-evident.