So after talking with a pretty amazing reader today. I was brought to some interesting conclusions. The first of these.
1. Maybe Corporate Genomics is not so bad after all.
I know...after all this railing.....you must begin to say. "Is this Sherpa crazy?"
No, this is a critical assessment. What is the risk of using these services? Discrimination? Well, health insurance discrimination is a fallacy. Very few employers have discriminated genetically.
Changing a behavior to a more deleterious behavior? Well, the McD's fries might get eaten more often if you don't think you are going to get heart disease is a reality. Will this set of tea leaves be perceived as strongly as a physician's interpretation? Or will this get treated like it is......a suspect set of data. That is the real question. If the answer is that the public treats the computer the same as the trained professionals, then we will have true risk. This has not been proven yet. Nor has the benevolence of these companies. But I can't say guilty until proven innocent. Especially because I have 10 readers from Mountain View daily (on average)
2. Maybe Primary Care Physicians can learn to perform genetics
Huh? I know. I have data which states that it takes more than a year to learn these topics. But, I don't have data for year 2 yet. Maybe in year 2 we see a tremendous uptick in knowledge and skills. Maybe not. But, I cannot make sweeping claims without this data.
3. Maybe Google, Kleiner Perkins, Genentech, NEA are in this for the benefit of mankind.
Ok, now I am strecthing. But maybe they are? Google has been investing in earth-friendly companies, education companies....KP is investing in Personalized Medicine, We know how Genentech has revolutionized some parts of medicine.......
You see. Socrates is correct. We need to challenge that which is taken as fact.
The Sherpa Says: Doctor Oliver Wendell Holmes said "Man's mind stretched to a new idea, rarely takes it's original form" Maybe I am wrong.....Maybe I am not.....But you have to take both as possibilities.
Friday, November 30, 2007
So after talking with a pretty amazing reader today. I was brought to some interesting conclusions. The first of these.
Posted by Steve Murphy MD at 4:33 PM
Thursday, November 29, 2007
Wednesday, November 28, 2007
Amazing today I have had to make the time for a second post. I received an email from Heather L. Shappell, M.S., CGC. She is the founder and COO of Informed Medical Decisions, Inc. The newsworthy event is that they have partnered with Aetna to delivery informed consent for genetic testing for cancer.
What is this company? Well......
From the site:
Informed was created to increase access to experts in cancer genetics for people at risk for hereditary cancer. Genetic counselors help people and their doctors make the most informed genetic testing, cancer screening, and cancer prevention decisions
This is precisely what they are now doing.
From the press release:
Aetna (NYSE: AET) today announced that effective immediately it will offer members confidential telephone and web-based cancer genetic counseling services as a component of health benefit plans which include coverage for genetic testing. The services will be offered through Informed Medical Decisions, a national genetic counseling company staffed with board-certified genetic counselors with expertise in cancer genetic counseling.
The GTO has also just put this on their post. I feel bad replicating data, but there are some people who may not get the GTO newsletter. If you don't, your missing out.
The Sherpa Says: Telegenetics has been around for a while. The Military has been using it and there are places such as Harvard which are trying out video genetics. It is a shame that their are too few people in the field. I think that Myriad is actually helping out physicians with this task. However, a word of caution. According to my research.....the more confident a non-genetics physician is at counseling, the less likely they are to do well on a genetics knowledge test. This is where Informed comes in....at least for cancer counseling....To do more than that you need more
Although this sounds great it is too bad Aetna has one of the most retrictive testing policies for BRCA out there....
Posted by Steve Murphy MD at 12:41 PM
Saturday, November 24, 2007
Thursday, November 22, 2007
Posted by Steve Murphy MD at 6:46 AM
Monday, November 19, 2007
Sunday, November 18, 2007
From Wired Magazine, a quote from Anne Wojcicki regarding her Gene Journal and risk calculator
"A lot of this is unknown. It's totally experimental," Wojcicki told me a few weeks before the science board meeting. "No one has looked at all eight diabetes markers together. They've all been identified individually, but they don't know exactly how they work together. So we've tried to make that clear."
To crunch these numbers and determine one person's risk factor, 23andMe has opted to multiply the risks together. But a competing school of thought argues for adding the risk from SNP to SNP. The two approaches can result in wildly different tallies.
Welcome to the first Google driven experiment in genetics, paid for by the customers......
Unfortunately that poor author from the NYT demonstrated her clear lack of that understanding. And she is a learned journalist, what's to happen to the layperson who has never had any education in genetics/science? Well, I guess you have to break a few eggs to make an omelet....... Primum Non Nocere....oh wait, I forgot, Anne isn't a physician...
The picture here is that of the Tuskegee syphilis study. A horrible experiment carried out on unwitting subjects that sparked the founding of Institutional Review Boards. IRBs were formed to protect research subjects. Why could Dr George Church only get a select few for his first personal genome project? The IRB deemed laypeople unable to give informed consent. How did they end around the IRB?
23andMe's mission is to be the world's trusted source of personal genetic information
Saturday, November 17, 2007
Today I plan to discuss where this magical field of personalized medicine is headed. A few things have changed over the week, and I think that this may change what I was originally going to discuss. But all in all the likely paths of personalized medicine could be up to 8-fold. Some are less likely and are therefore not discussed here.
In order to really understand personalized medicine's future we must know 4 truths.
1. Genotyping is getting cheaper and cheaper. Less than a penny a base pair. So in turn genetic testing should get cheaper and cheaper. This is the idea of a 1000 USD whole-genome. Unfortunately some tests still remain in the thousands of dollars. Therefore, I assume (unless these companies are paying their workers millions), that there is some significant money being made here.
2. There is a lack of healthcare specialists trained in the field of genetics. In fact the healthcare system fails all of us when it comes to genetics education and understanding.
3. The alternative healthcare market is skyrocketing. Do-it-yourself (DIY) treatment with herbs, vitamins, and nutraceuticals is predicted to be a trillion dollar market in the next decade. As a corollary, because genetics is not well understood by the "standard" medical practitioner, many view this as an alternative healthcare style
4. Insurance reimbursement for even standard genetic tests and counseling is woeful. If you are with an insurer such as Empire BCBS coverage of a BRCA panel is limited only to those who have breast cancer, or those who have 3 family members with such cancers at frighteningly young ages. This is better than others, who flat out refuse to cover BRCA testing unless a letter of medical need is written(taking valuable time from genetics providers).
This leads me to the paths. I think the paths are reliant on 3 things
1. Legal issues such as GINA approval and legal requirements changing for genetic testing
2. Public opinion for genetic technology in medicine remaining favorable
3. Personalized Medicine not being oversold
Probable Path 1
GINA is passed and states continue to have strong laws regarding genetic testing .
In this environment of protection from discrimination, polls have already demonstrated that the public will continue to view genetics as good for medicine. Therefore there will be a market for personalized medicine services. Private and public investment will continue to fuel the machine. However,there will be significant legal challenges to non-medical providers offering genetic services. But, personalized medicine will thrive. Discovery will continue and the public will continue to be excited. Personalized medicine will start to be oversold and the public may go wary. We are already seeing these trends with companies such as Navigenics. If lawsuits should happen, and they will, private investors will back away from this potential landmine
Probable Path 2
GINA dies in the Senate and strong laws regarding genetic testing continue to be enforced.
In this environment the public will still continue to be wary about discrimination. Most will not see value in giving a company, not governed by HIPAA, their genomic information. This is the environment we currently are in. All of this investment in these genome sequencing companies can go awry if customers’ personal health information can be sold to marketers looking to personalize ads. In addition if the government wishes to investigate DTC companies, as it is currently doing in CT, private genome sequencing could be a fad like Z-Cavaricci’s. Cool to have for a little while,until you realize that they put you at risk of being ridiculed or discriminated against. Now smart ad people will tell you that it is worth the risk, they may even tell you “Why not have the data now? You can use it later, but why not buy it now? Everyone is doing it.” Until that first case of genome data showing up when your prospective employer googles you. I imagine that would be worse than those “party” photos you still have up on your MySpace site.
Probable Path 3
GINA is passed and laws relax on genetic testing
In this sweet spot for corporate testing we see the rise of consumer genomics. The public opinion views this technology as benign as pregnancy testing. This is a scenario where the movie GATTACA takes place. “Genetics as a tool for better living”. Fueled by investment the future of personalized medicine is strong. We will see sex selection, PGD for athletic abilities. Since employers cannot discriminate based on “genetics” we may see some lawsuits because of genetics being used as a surrogate for “abilities” Because of the lack of legal consumer protection, the overselling may end up killing the best parts of this field. The databases of genomic data for research will be overflowing as patients will have no fears. IRBs will not have to treat genetic data and differently and perhaps restrictions on genetic research will be reduced. Discovery could be enhanced in this environment. I can see why many PhDs would favor this scenario.
The Sherpa Says: There are many more paths mapped out in my mind. I, the Sherpa have envisioned this mountain for decades. This is my passion. My existence. Maybe you should ask your friendly neighborhood consumer genomics company if it's theirs too. One things for sure, overselling via the NYT, Forbes and 23andMe will kill any chance of personalized medicine being viewed as a credible, respectable, medical specialty. This could be the beginning of the end of personalized medicine!
Thursday, November 15, 2007
Tuesday, November 13, 2007
Lo and Behold, the Sherpa goes looking for some updates, Epidemix and Wired deliver. My favorite line from Wired is
"Early medical testing and treatment could save patients and healthcare providers a ton of money, but nobody wants to pay for unproven and often expensive new lab work. FDA approval is not required for laboratory tests, but it is an indicator that products are actually beneficial to doctors and patients."
The best lines from Epidemix are
We’ll see about that - but there were three telling stats that came up during the day. Together, they make quite the case for personalized medicine.
1) Half of all prescriptions don’t work for the patients. Most drugs have an efficacy between 20 and 80 percent, averaging around 50 percent. Meaning that they only have their intended effect half the time. That might be awesome in baseball, but it’s hardly reassuring in medicine.
2) Chemotherapy is effective - defined as remission - in just 5 to 10 percent of breast and colon cancer cases. This is likewise startling (the stat comes from Randall Scott of Genomic Health). And factor in the fact that chemo costs about $30,000 per patient per year, and there’s a massively inefficient treatment module out there.
3) Six weeks - that’s how long it takes, give or take, for a physician to determine whether a given antidepressant is working for a patient. And given that only half of drugs work, that’s a rather long time for a patient to go effectively without a treatment for their depression or mental illness. (This from Wolfgang Sadee, chair of the pharmacology department at Ohio State).
But what is most exciting about the conference is the attention that was purported to be spent on predicition, prevention and personalization. This is what the essence of personalized medicine is.
Personalization includes face to face care and the ability to ask questions. Not questionnaires that give you incorrect guidance. No amount of computer work can give you the face to face communication. Here is an example of preposterous questionnaire results.
If I was a 40 year old woman with no first generation history of breast cancer, but had a more distant relative with breast cancer after 50, HH's position and that of the literature is that you probably shouldn't have BRCA testing as the likelihood is very low.You may ask why and spend time with us going over why we think so. But in a questionnaire the answers are often yes or no. Much like the one I just filled out. Don't believe me? Try it yourself
Direct sequencing is probably not the best "test" to detect and prevent breast cancer, a mammo is. And it certainly is tricky how this info comes up first, rather than the next screen where it says this test may not be appropriate for you. A questionnaire is only as good as the answers they provide....
The Sherpa Says: In the scramble to make money off of genetics the consumer should make sure they get what they pay for. If the cost is cheap, it probably means the same for the service....... I hope the public can understand that. Thanks to Wired and Epidemix for the great posts. I look forward to hearing the rest of the conference tid bits! Oh, and shame on those little questionnaire writers.....Did you really think someone with genetics training wouldn't pick up on that?
Monday, November 12, 2007
Sunday, November 11, 2007
"If we could merge the real advantages of these companies:
- the fantastic team of Navigenics and their unique business model;
- the financial background of 23andMe; the focus on genealogy information and social networking;
- the personal aspect of Helix Health of Connecticut and their potential to serve and help physicians as well,
Friday, November 9, 2007
Hot off the presses atthe American Heart Association conference, results of a geneotypic based coumadin dosing strategy were released. The results of this 200 person study were pretty neat.
"Out-of-therapeutic-range prothrombin times were not less common with the pharmacogenetic algorithm than with standard empirical dosing, But initial doses chosen based on genotype were closer to patients' eventual stable doses (P<0.001) and subsequent dosing adjustments were smaller (P=0.002) and less frequent (P=0.03) than with empirical dosing, reported Jeffrey L. Anderson, M.D., of Intermountain Healthcare and the University of Utah in Salt Lake City, and colleagues."
"The Couma-Gen study included 200 adults with an indication for warfarin who were randomized to receive the drug according to a standard algorithm of 10 mg on days one and two followed by 5 mg daily or by a pharmacogenetic algorithm based on genotype, age, sex, and weight."
Most notably a subgroup analysis (which is suspect based on the small sample size) found
"Pharmacogenetic-guided dosing failed to reduce the percentage of INRs per patient that were outside the therapeutic range compared with standard dosing (30.7% versus 33.1%, P=0.47), which was the primary endpoint. However, there was a significant benefit when excluding patients with only one variant (41% of patients overall)."
So where does all of this dosing decision go from here? I await the CROWN study results as well as what Washington U St.Louis has to say.
Lastly, I have had only one comment on my last post. Why do you think DTC testing is Illegal in NY and CT. Please read this post and make a comment! The last one was pretty interesting. I wonder if he's tried do it yourself surgery yet ;)
Posted by Steve Murphy MD at 8:14 AM
Thursday, November 8, 2007
Tuesday, November 6, 2007
“…. Any laboratory test of human DNA, chromosomes, genes, or gene products to diagnose the presence of a genetic variation linked to a predisposition to a genetic disease or disability in the individual or the individual’s offspring; such term shall also include DNA profile analysis. ‘Genetic test’ shall not be deemed to include any test of blood or other medically prescribed test in routine use that has been or may be hereafter found to be associated with a genetic variation, unless conducted purposely to identify such genetic variation.”
For clinical genetic tests, the informed consent must state that the sample shall be destroyed at the end of the testing process, or not more than sixty days after the sample was taken, unless a longer period of retention is expressly authorized. New York law requires individual authorization for sample retention for up to ten years if no genetic testing is performed; however, informed consent must be obtained prior to the conduct of genetic tests. Retention of a DNA sample past a period of ten years requires explicit consent for a longer or indefinite period of retention.
Laboratories may accept specimens only upon request of licensed physician or other persons
authorized by law to make diagnoses. Laboratories may report findings only to the licensed provider that ordered the test. Laboratories may provide results to lay persons upon written request of the provider who ordered the test. An official at the Connecticut
CLIA Laboratory Program confirmed that DTC testing is not permitted.
N.Y. Pub Health Law § 576-b N.Y. Pub. Health Law § 577 10 NYCRR § 19.1(j) 10 NYCRR § 58-1.7 10 NYCRR § 58-1.8 10 NYCRR § 63.3(e)
In general, tests may be ordered only by licensed physicians “or other persons
authorized by law to use the findings of laboratory examinations in their practice or the
performance of their official duties.” Consumers are not listed among those
authorized. Test results cannot be sent directly to patients except with written consent of the
physician or authorized person, except blood type and RH factor can be given in writing to
the patient without written consent. DTC testing is permitted for tests that have been
approved by the Food and Drug Administration for direct, over-the-counter sale
DTC testing is not permitted, other than for certain tests relating to the blood supply, such as HIV and Hepatitis C tests.
When reviewing the web today it has become clear that the media is again hyping things today.
A study designed to evaluate the effects of ONE GENE on the role of IQ development in breastfed vs non babies. At first read does this sound like it could be true? I mean does the entire neurodevelopment of IQ hinge on this one gene.
Hsien Lei covers the story over at EyeOnDNA and so does Reuters.
Why is this story fishy at best?
1. According to the authors "We took cells from the children and then analyzed DNA and then we compared how they scored on IQ tests and looked up if they were breast-fed as babies," "It was very straightforward." Retrospective analysis is genetics is never the gold standard! Prospective is.
2. The authors even state (IN THE ARTICLE) that the modification of IQ is not likely to be due solely on this one SNP!
3. This is the FIRST study and replication is essential. But where do you read that at Reuters?
The Sherpa Says: That being said, this could be a pathway. I bring to your attention the age old debate about coffee being good/bad for your heart. When looking at environmental effects through a genomic eye the answer was found. Coffee is good if and only if you process it fast enough. So maybe we should be checking out FADS2 and CYP 1A2.... Let's not jump the gun just yet. Maybe the gene is just a marker and has nothing to do with breastfeeding at all. Just like this previously linked story....
Sunday, November 4, 2007
Thanks again to Bertalan Mesko for putting the Sherpa on the Map at the Gene Genie. Number 19 is up at his blog Scienceroll. A nice Texan tells us about our 95% "Junk" and the 0% in the Puffer Fish.
More importantly today I read a verification of what I already know. The United States population is fed up with our healthcare system. Is that a surprise to anyone? I know that the patient is fed up with short visits, uncoordinated care, and increasing costs. Why pay for a system that fails to place emphasis on preventative care? I for one am fed up with the system we have.
But what is the alternative? As a farily recent medical grad I face 280,000 USD in debt. In addition, I am in a field where the governmental rate at which the pay for interpretation of a complex genetic test is 18 USD. How many tests would I have to interpret to pay that cost back? You do the math. This is precisely why you have so few genetics providers in the US. The same is true for primary care physicians. We are headed down a dangerous road.
How will we ever achieve personalized medicine if the physician only has 20 minutes to do your preventative care. Lipids, Blood Pressure, Glucose, Colonoscopy, Mammo, Pap Smear. Once I address all of these things there is no time to take the family history. There is a reason why most EMRs don't have good family history intakes....There is no CPT code for taking one!! If you don't get paid for it, some would argue "Why do it?"
With Thanksgiving coming up I would like to remind everyone of the Department Of Health and Human Services tool to take your own family history. Take this time on this national family history day and use this. It truly could save your life.
The Sherpa Says: If Medicare pays 18 dollars for test interpretation (The largest amount of any insurer).....Maybe we should ask the public what they think it is really worth.
Posted by Steve Murphy MD at 11:26 AM