Obama and Genomic/personalized medicine

[HESC]

So, Obama brought forth some legislation called the Genomics and Personalized Medicine Act back in 2006, another senator brought forth a revised version of the bill this year. Now Obama has apparently said he would enact some of the bills provisions through executive order to make the CDC and HHS make genomic and personalized medicine happen.


Does anyone think this will really work for one thing, and second, is personalized medicine what pathologists have been waiting for? Are your attendings bantering as much about this as mine are?</O
<O</O
<Ohttp://www.govtrack.us/congress/bill.xpd?bill=s109-3822&tab=summary

---

Members don't see this ad.

 

[Gut Shot]

I certainly hope it works... a lot of institutions are investing a lot of money pursuing it as the next great thing. Baylor Houston is building a new frickin' hospital that revolves around this concept.

In my own practice I'm just expecting to get progressively buried under more and more requests for point mutation testing.

 

[HESC]

In my own practice I'm just expecting to get progressively buried under more and more requests for point mutation testing.

I guess point mutation testing isnt profitable?? or interesting?

 

[pathgrrl]

I guess point mutation testing isnt profitable?? or interesting?

Molecular testing is not profitable, IMO, it is labor intensive and at this point poorly automated. Also, many assays require paraffin shavings so that's a lab-to-lab transfer problem with blocks sometimes requested from the outside pathology lab.

 

[Gut Shot]

I guess point mutation testing isnt profitable?? or interesting?

The situation right now is largely irrelevant. The question is how things are going to look in 5, 10, 15 years, and so on. As a prelude to what we may be in for, I would point to K-RAS mutation testing (mutations that predict therapy response in various carcinomas).

Combine this with announcements like this one, where a genomics company is going to sequence 1,000 human genomes in 2009 for $5,000 apiece, and perhaps you're starting to get the idea.

Now, all that said, how much clinically relevant information will be revealed by genomics is anyone's guess. I suspect it's going to be a lot, but we shall see.

 

[LADoc00]

Mkay, here is where I get fairly pissed.

things like K-ras, B-raf and Oncotype Dx mutational analysis are NOT personalized medicines folks.

It is stratification of the same herd data you get from things like ER, PR, ki-67and EGFR by IHC.

All these tests look at how someone with trait X such as K-ras wild type would typically respond, it is still a PROBABILITY analysis.

The fact you are using PCR to look at the DNA doesnt make it more personal than simple IHC tests we already do. This is sham operation being sold by a massive flood of biotech start up companies to investors and consumers alike. It is scam and I dont care how many BS articles are written about this.

Personalized medicine would be analyzing your DNA and using some type of combinatorial chemistry-like method to test response in accurate in vitro models of YOU (such as cell lines or supercellular structures).

We are decades if not a century away from this.

Our scientific leadership and politicians in this country are basically idiots.

 

[Gut Shot]

things like K-ras, B-raf and Oncotype Dx mutational analysis are NOT personalized medicines folks.

It is stratification of the same herd data you get from things like ER, PR, ki-67and EGFR by IHC.

All these tests look at how someone with trait X such as K-ras wild type would typically respond, it is still a PROBABILITY analysis.

The fact you are using PCR to look at the DNA doesnt make it more personal than simple IHC tests we already do. This is sham operation being sold by a massive flood of biotech start up companies to investors and consumers alike. It is scam and I dont care how many BS articles are written about this.

Personalized medicine would be analyzing your DNA and using some type of combinatorial chemistry-like method to test response in accurate in vitro models of YOU (such as cell lines or supercellular structures).

We are decades if not a century away from this.

Yeah, that would take awhile, but it has no bearing on what is actually happening. The IHC examples you listed above are indeed personalized medicine, albeit rudimentary. Using biomarkers to tailor therapy for a specific patient... what's not personalized about that? According to the Personalized Medicine Coalition (which includes everyone from the FDA to Harvard):

We define personalized medicine as the use of new methods of molecular analysis to better manage a patient's disease or predisposition towards a disease. It aims to achieve optimal medical outcomes by helping physicians and patients choose the disease management approaches likely to work best in the context of a patient's genetic and environmental profile.

You'll note the use of the word "likely" in the above quote. Anyone with a basic understanding of genomics (which is what current goals are centered around) will appreciate that probability analyses are inescapable. If it works out, and to paraphrase Jim Goad, genomics will be to IHC biomarkers as sexual homicide is to a goodnight kiss.

 

[LADoc00]

the biomarkers are based on herd data from thousands of prior patients, there's nothing "personal" about it. Its merely racial profiling at the molecular level.

Looking at tumors via H+E and simply grading them is "personal" by that definition too!

In time when the healthcare $ is evaporating do we need more of this smoke and mirrors crap to drive Medicare to bankruptcy any sooner??

 

[Gut Shot]

the biomarkers are based on herd data from thousands of prior patients, there's nothing "personal" about it. Its merely racial profiling at the molecular level.

Yep, that's sorta the point: the herd gives you probability data to apply to the individual. You sound surprised.

Looking at tumors via H+E and simply grading them is "personal" by that definition too!

As is looking at them grossly. But you have to admit, the type and amount of information obtained between the two techniques is vastly different.

 

[HESC]

the biomarkers are based on herd data from thousands of prior patients, there's nothing "personal" about it. Its merely racial profiling at the molecular level.

Looking at tumors via H+E and simply grading them is "personal" by that definition too!

I think actually looking at the tumor/or whatever is way more personal than simply putting a patient into a diagnostic/treatment/management algorithm based on subjective symptoms. (how many normal appy's have you seen whose accessions said acute appy? = not personalized )
anyway, would you not agree that the more information about a pathology we can get the more refined the clinical management can be? We have come a long way from diagnosing only "blood cancer" and have personalized it quite a bit with molecular data

 

[HESC]

Bill Introduced to Modernize Lab Fee Schedule
http://www.g2reports.com/issues/news/news/breakingnews/484-1.html

anybody heard anything about this? are attendings/chairs/ PDs anticipating this bill going through? It claims that if they fee schedule would be adjusted for inflation only then labs would be reimbursed 25% more!!! how big of an impact would an increase of 25% make on the job market?? interesting possibilities...

 

[docbiohazard]

Bill Introduced to Modernize Lab Fee Schedule
http://www.g2reports.com/issues/news/news/breakingnews/484-1.html

anybody heard anything about this? are attendings/chairs/ PDs anticipating this bill going through? It claims that if they fee schedule would be adjusted for inflation only then labs would be reimbursed 25% more!!! how big of an impact would an increase of 25% make on the job market?? interesting possibilities...

Mmmm Change ($$$) I Can Believe In.. [tm]

Have to see what comes to pass...

DBH

 

[HESC]

This is from the President’s Council of Advisors on Science and Technology (PCAST) september 2008 report "Priorities for Personalized Medicine".

Priority Area 3: Reimbursement
Challenges

There are three key challenges to achieving cost-containment objectives for health care without arbitrarily obstructing the adoption of innovative genomics-based molecular diagnostics. The first challenge is that reimbursement of genomics-based molecular diagnostic tests as low-margin commodity items – as is common practice for laboratory diagnostics – will reduce the likelihood that such products will be developed by industry. The second challenge is the need to develop standards for the evidence that CMS and other payors will require
to validate the benefits of these tests in real-world settings. The third challenge involves the procedural hurdles associated with coding systems, bundled payment systems, and complex billing procedures and requirements that can especially impact reimbursement for innovative molecular diagnostics.​

Policy Recommendation
6. Public and private payors should determine coverage policies and payment rates for genomics-based molecular diagnostics in light of their overall impact on patient care, as demonstrated by evidence from clinical trials and other well-designed empirical studies.

• Public and private payors should reimburse for genomics-based molecular diagnostics commensurate with the clinical benefits provided and should collaborate with test developers to establish new, more flexible coding approaches for reimbursement.
• Public and private payors should collaborate to expand “coverage with evidence development” programs that extend coverage and reimbursement while a product is being investigated for appropriate use and effectiveness.​

• Public and private payors should collaborate in the development of standards for clinical trial designs that would be accepted as providing evidence sufficient for coverage decisions.

this report is another of many (IOM 2000, CDC more recently) that calls for the improvement of reimbursement of lab tests, (AKA reimbursed according to the clinical benefits). Looks like path might come out a winner from the impending health care reform...

 

[Gut Shot]

Here is an article from last November about what Baylor is attempting to do with their new, under-construction hospital. In pathology we tend to immediately leap on personalized medicine in the context of oncology, with the goal being to tailor prognostic information and guide therapeutic decisions. But the other half of the equation is determining lifetime disease risk, mainly for common conditions. Anyways, enjoy:

Baylor's Tweardy Discusses New Chip, Personalized Medicine Teaching Hospital

Excerpt:

Could you give me some details on the Baylor chip that you are developing?

The chip really covers five general areas, and all of it is Illumina based. There are 32 genes. The categories are pharmacogenetics testing; SNP testing for common diseases; a SNP-based assessment of HLA transplantation antigen; single-gene disorder testing for mutations in 99 genes that predispose adults to certain diseases; and a SNP profile to assess genetic or ethnic background.

Could you elaborate on the last category, SNP profiling to assess ethnic background?

The HapMap Project, which took 30 individuals from Western European background in Utah and 30 individuals from Africa, and 30 individuals from the Pacific Rim, demonstrate clearly that there are SNP differences that are just based on the ethnic background. In order to interpret SNP results that you get for the other four categories, particularly the common risk diseases and the HLA transplantation antigen, which are SNP-based assessments, you need to put those into an ethnic context in order to interpret them correctly. That's an important component of that chip now. It's somewhat controversial because ethnic profiling is not something that you can enter into lightly. You have to be careful what you do with that information and all of that. But it turns out it's critically important to interpret the SNP analysis for the other tests that we're doing.

You have said you plan to introduce the chip first in your women's health clinic. How will you use the chip in the clinic?

We're going to use the chip to put together a prevention and management plan for those individuals who come into the women's health clinic. It will be used for research but in fact the immediate benefit to the patient will be that we'll be able to help manage their healthcare better with the information we get from the chip. It will be immediately beneficial in terms of alleviating patient worry, changing lifestyle, and also managing medications they are on. The plan would be to offer it to these women for the purposes of helping to manage their healthcare.

In the way that you describe the chip, it doesn't seem particularly specific for women's diseases. Why are you introducing it in the women's clinic first and are there other populations you intend to introduce the chip in?

We're targeting it to the women's health clinic because we have a very interested and probably a little more financially well-off population that would like to have a level of care and would be willing to pay for that additional level of care. As you know, these chips are not reimbursable for general use. So, that partly influenced our decision with where to go. We wanted to go into an interested subpopulation of patients that would see the benefit, the worth of it, and who could also pay for it. And the group that we have set up right now in that category is those in our women's health clinic.

 

[SaltySqueegee]

Personalized medicine from the stand point of combinatorial chemistry is a very long ways off (agree with LA). Chronic complex genetic traits which set in at late-stage cannot be simply tested using an assay of this sort, because they intrinsically involve multiple pathways with subtle low-penetrant changes. You will not pick this up in a simple cell-culture or colorimetric assay. Analytical CV will be higher than biological CV for the cellular pathways tested.

This is the issue at hand folks. And don't get me started on genome wide association studies... to find risk loci. The odds ratio for individual changes are not clinically relevant.

-Salty.

Mkay, here is where I get fairly pissed.

things like K-ras, B-raf and Oncotype Dx mutational analysis are NOT personalized medicines folks.

It is stratification of the same herd data you get from things like ER, PR, ki-67and EGFR by IHC.

All these tests look at how someone with trait X such as K-ras wild type would typically respond, it is still a PROBABILITY analysis.

The fact you are using PCR to look at the DNA doesnt make it more personal than simple IHC tests we already do. This is sham operation being sold by a massive flood of biotech start up companies to investors and consumers alike. It is scam and I dont care how many BS articles are written about this.

Personalized medicine would be analyzing your DNA and using some type of combinatorial chemistry-like method to test response in accurate in vitro models of YOU (such as cell lines or supercellular structures).

We are decades if not a century away from this.

Our scientific leadership and politicians in this country are basically idiots.

 

[HESC]

Personalized medicine from the stand point of combinatorial chemistry is a very long ways off (agree with LA). Chronic complex genetic traits which set in at late-stage cannot be simply tested using an assay of this sort, because they intrinsically involve multiple pathways with subtle low-penetrant changes. You will not pick this up in a simple cell-culture or colorimetric assay. Analytical CV will be higher than biological CV for the cellular pathways tested.

This is the issue at hand folks. And don't get me started on genome wide association studies... to find risk loci. The odds ratio for individual changes are not clinically relevant.

-Salty.

The NewEngland Journal of Medicine disagrees with you! A pharmacogenetic study (which IS personalized medicine) came out in feb this year showed that using genetic testing for variations of CYP and VKOR made a big difference in acheiving INR on the initial dose of warfarin when used with the standard clincal algorithm. they concluded the following:

The use of a pharmacogenetic algorithm for estimating the appropriate initial dose of warfarin produces recommendations that are significantly closer to the required stable therapeutic dose than those derived from a clinical algorithm or a fixed-dose approach. The greatest benefits were observed in the 46.2% of the population that required 21 mg or less of warfarin per week or 49 mg or more per week for therapeutic anticoagulation.

The article also alludes to a prospective clinical trial to determine if this test actually leads to improved clinical outcomes. which, if you have ever seen a patient in the ICU with a brain/GI blead due to warfarin toxicity it is seems like it should lead to a more costeffective practice. Personalized medicine is here and pathologists should recognize the value of this new trend and embrace it instead of wasting the opportunity to increase their scope of practice. you can be a contrarian if you so like but you might find that your clinical colleagues turn to someone else when they need a consult of this nature.

 

[SaltySqueegee]

Warfarin toxicity susceptibility is one of the FEW/RARE traits that is MENDELIAN. The world of clinically relevant Mendelian traits has already been tapped. Yes, this test is useful, however you need to realize that Mendelian traits is not what Molecular Diagnostics is struggling with right now. We are at the cross-roads of dealing with MULTI-FACTORIAL Complex Genetics that deal with a combination of rare loci, and common polymorphic and structural genetic changes that are in unique combinations to each individual, and it is HIGHLY unknown at this present state how they all fit together.

For instance. BRCA mutations account for 10-20%% of breast cancer cases as a susceptbility loci. It is highly penetrant and is considered Mendelian in many ways. However, the rest of the individuals, 80%, that are diagnosed with breast cancer in their lifetime will not have a simple single loci explanation. It certainly would be nice to help the other 80%. But it's a long ways off.

One goal for personalized medicine is to identify those individuals at risk ahead of time so that they can be closely monitored, enrolled in clinical trials, etc. Anything to reduce the false positive rates and increase the power of the studies. Identifying the mechanistic basis for breast cancer risk in the rest of the 80% that don't have BRCA mutations is not feasible right now. It's not that I'm not willing to embrace personalized medicine, its just that there are literally not enough people in the world to reach statistical signifcance and power for studying these complex disease traits. This is where we're at.

A new way of approaching complex trait studies needs to happen before personalized medicine will take off.

Read Nature reviews genetics for some dialog on the topic.

The NewEngland Journal of Medicine disagrees with you! A pharmacogenetic study (which IS personalized medicine) came out in feb this year showed that using genetic testing for variations of CYP and VKOR made a big difference in acheiving INR on the initial dose of warfarin when used with the standard clincal algorithm. they concluded the following:

The article also alludes to a prospective clinical trial to determine if this test actually leads to improved clinical outcomes. which, if you have ever seen a patient in the ICU with a brain/GI blead due to warfarin toxicity it is seems like it should lead to a more costeffective practice. Personalized medicine is here and pathologists should recognize the value of this new trend and embrace it instead of wasting the opportunity to increase their scope of practice. you can be a contrarian if you so like but you might find that your clinical colleagues turn to someone else when they need a consult of this nature.

 

[HESC]

We are at the cross-roads of dealing with MULTI-FACTORIAL Complex Genetics that deal with a combination of rare loci, and common polymorphic and structural genetic changes that are in unique combinations to each individual, and it is HIGHLY unknown at this present state how they all fit together.

I must contend that this kind of assay IS NOT FAR OFF. currently the FDA has approved 3 in vitro diagnostics multivariate index assays (IVDMIA) (see mammaprint, oncotpyedx, allomap) that use multiple genes and algorithms to predict dz recurrence, response to chemotx and likelihood of transplant rejection. There has been such a surg of intereset in producing these type of assays that the FDA has taken interest in regulating them (see the pdf of FDA draft guidelines for IVDMIA or the OIVD home page ). this IS NOT far off, IT IS HAPPENING RIGHT NOW!

For instance. BRCA mutations account for 10-20%% of breast cancer cases as a susceptbility loci. It is highly penetrant and is considered Mendelian in many ways. However, the rest of the individuals, 80%, that are diagnosed with breast cancer in their lifetime will not have a simple single loci explanation. It certainly would be nice to help the other 80%. But it's a long ways off.

if you are talking about sequencing someones/an infants genome to predict what will occur to them when they are 50-60 yo then i would agree with you that is a not going to happen for some time. but like i have pointed out there are assays that are out on the market today that represent the power genomic technology and personalized medicine. It is, of course, in its infancy in terms of a mature technology but that does not mean true applications are far off.

One goal for personalized medicine is to identify those individuals at risk ahead of time so that they can be closely monitored, enrolled in clinical trials, etc. Anything to reduce the false positive rates and increase the power of the studies. Identifying the mechanistic basis for breast cancer risk in the rest of the 80% that don't have BRCA mutations is not feasible right now. It's not that I'm not willing to embrace personalized medicine, its just that there are literally not enough people in the world to reach statistical signifcance and power for studying these complex disease traits. This is where we're at.

while we cant predict who will get breast cancer in the remaining 80% of pts, we can stratify those patients with breast cancer to determine the optimal therapeutic approach (see oncotypedx publications , it appears this assay is much better than mammaprint.) And if anyone is thinking that this is unlikely to catch on, see this study from kaiser. (kaiser is probably going to be the model for how all community hospitals are run within our careers b/c they are profitable and delivery a decent quality of care). anyway, it is likely that this kind of strategy is cost effective and provides the right treatment for the right patient instead of over/under treating everyone and dealing with the complications/costs, which is what will probably determine what happens in medicine with the impending reform/cost crisis. again, this IS personalized medicine.

Read Nature reviews genetics for some dialog on the topic.

indeed, Nature reviews is an excellent resource for basic science. I must refer you to read the New england journal of MEDICINE, they have good dialog on whats actually going on in medicine. (for oncotypedx A Multigene Assay to Predict Recurrence of Tamoxifen-Treated, Node-Negative Breast Cancer)

But good debate Salty! hopefully i got my point across and that we shouldnt have to keep convincing people in the pathology forum that this technology is usefull. we should instead turn our attention to convincing and educating the clinicians about this sort of stuff so that they will actually order the tests and the real world trials can take off.

 

[SaltySqueegee]

The reason why the FDA stepped in is because these tests have such poor inter-lab correlation. I can speak specifically of the mammaprint agendia assay from Amsterdam. When the lab I worked in submitted our 510K investigation for a new diagnostic device to the FDA we used the agendia application for designing our case-control studies and powering for the test we submitted. The FDA's time is precious, and we wanted to make sure we modeled portions of our app after a successful one. The problem with agendia, which we had to control for in our studies, was making sure the appropriate standards were in place to control for inter-lab and inter-assay variability that agendia suffered from, and still suffers from. After reviewing the agendia FDA clearance they only let them perform the diagnostic test in the lab that performed the initial case-control studies: Amsterdam! It's not that the assay doesn't reproduce, it just doesn't reproduce in other clinical labs. That is a problem. So, despite the fact we have this wonderful test, only one lab overseas is FDA cleared. This is not high throughput, cheap, nor will it be the model for personalized medicine in the future.

The problem is over-application of microarrays. Microarrays are wonderful tools for discovery, but suffer analytically in the realm of lower limit of quantification, linear dynamic range, false positives, false negatives and poor inter-lab correlation. See Microarray Quality Consortium studies in Nature Biotech 2006, Canales. Agendia discoverd their multi-gene index marker on a microarray platform, and made the crucial mistake of sticking with a microarray platform as the diagnostic to submit to the FDA. The FDA will not clear them until they can demonstrate that they have cleared this hurdle.

The in vitro "PCR" based diagnostics that are FDA cleared and have excellent inter-lab reproducibility are all based on "competive" PCR methodology (HIV assay by Roche comes to mind). Competitive being the inclusion of competitive-internal-standard templates in PCR reactions that allow for inter-lab quality control. Microarrays are not amenable to this modification (intrinsic problem with the method). Developing an assay based on these quality control standards takes YEARS. Believe me, I know personally.

So when I say it will be a long time before we see personalized genomic medicine to take hold in a strong fashion, it is 1) because of the complex nature of it (see prior posts), and 2) because bench to bedside development and validation of diagnostics takes a huge amount of time; especially if you want to use the assay in multiple labs in easy to use kits (not easy).

I can't comment personally on the other assays you cite, but if they are based on microarrays or real-time PCR, they will still suffer from the same analytical performance problems (real-time primarily inter-lab reproducibility, and not the linearity issues with microarrays). Biotrove, however, has adapted the competitive templates to their real-time platform, so I suspect discoveries made using real-time may be easily transferred into the diagnostic setting in the future. This still takes some work, and this recent event is still in the R&D stage.

 

[HESC]

Developing an assay based on these quality control standards takes YEARS. Believe me, I know personally....

.... I say it will be a long time before we see personalized genomic medicine to take hold in a strong fashion,

#1) agreed. we look towards you and your md/phd to take on that process and come up with some good stuff! I'll keep working on stem cells
#2) i believe you are correct with genomic assays especially taking a little longer, but there is real progress going on especially on highthroughput sequencing and PCR technology (including methylation assays).

personalized medicine is here today however, in the form of her2/neu, EGFR/Kras testing to determine therapeutic response, the warfarin test i cited in a previous post and etc...

and some of the lab interreproducability problems you spoke of are indeed problems, however, many diagnostic tests in use have problems. if you think of POCT you think of problems with the devices giving results different than the core chemistry analyzers. Her2/neu testing has problems but that hasnt stop it from making a huge impact on cancer care. Anatomic path suffers from inter and intra observer variability but that hasnt stoped AP from being the gold standard. thats just the nature of testing...
but i would be interested in hearing how you are getting around these problems in your own research because i think it is important.

 

[HESC]

i think we have sort of devolved into to detailed of an analysis for this thread.

however it is very interesting

 

[HESC]

an excerp from the genomeweb.com's pharmacogenomics reporter

Investment Bank Reports Advantage of CLIA Labs in Driving Adoption of OncotypeDX

Genomic Health's marketing and education efforts appear to be fruitful. The company netted $110.6 million in revenues last year, compared with $64 million in 2007. It delivered 39,600 test results in 2008, a 62 percent increase from the year-ago period. As of 2008, the company estimates that more than 90 percent of the node-negative breast cancer patient population is covered for the test.

looks like this test is utilized more than I thought...

the article also talks about the FDA regulation of IVDMIA and the need for the FDA to mandate testing be done before giving erbitux and such.

it also states that genomic health (the company that makes oncotype dx) is creating similar assays for DCIS and CRC. interesting...

 

[BrainPathology]

the biomarkers are based on herd data from thousands of prior patients, there's nothing "personal" about it. Its merely racial profiling at the molecular level.

Looking at tumors via H+E and simply grading them is "personal" by that definition too!

In time when the healthcare $ is evaporating do we need more of this smoke and mirrors crap to drive Medicare to bankruptcy any sooner??

It's true that all of these tests are "personalized medicine", just as all medicine has always been in all of history. We have constantly tried to more specifically target treatment to a patient's condition.

It's also true that the public is being scammed and mislead that this particular type of progress is something new or revolutionary. As if doctors up until now took nothing at all about their patient's unique situations into account. Reading the hype about the next new shiny molecular tests would certainly have you believe that though.

We SHOULD be calling this out just as LaDoc is! (dear lord, but I wholeheartedly agree with him on this... ). The most obvious to me though there may be plenty of more obvious ones is that we've had "personalized medicine" in the form of antibiotic resistance testing for years.

In pathology specifically there is a laundry list of tests that have been around for varying numbers of years to decades that meet the test of "personalized medicine." There is an immunohistochemical marker for INI-1 mutations in AT/RT. There is an immunohistochemical marker for IDH-1 mutation in gliomas. There are immunohistochemical markers for stratification of medduloblastomas. Hell in the right pathologists hands H&E is just as good a "marker" for 1p/19q deletion in oligodendrogliomas as FISH (with stringent enough criteria the agreement with histology and FISH approaches the high 90's %). For all intents, there's an H&E marker for t(X,17). The immuno markers for breast cancer have been mentioned several times already. There's an H&E marker for t(1,13) or t(2,13), but the exact same treatment you'd get for both mutations, stage for stage, is somehow "personalized" when the doctors spend the extra K$ to find out which one it is (it's prognostic..that's another argument that I don't have too many strong feelings about).

There's been fits and starts and occasional leaps and bounds made in this form of "personalized medicine" forever. It's ridiculous to now claim a revolution is coming, or that we need the most expensive most new, most technically difficult test we can think to run to get to the threshold of "personalized." When people discover this scam it will most likely hinder progress of new difficult novel tests that DO matter, like some of the ones about to come on line above.

 

[Tissue issue]

From Wikipedia-
According to Anne Wojcicki, 23andMe has been in dialogue with the FDA since 2008.[23] In 2010 the FDA notified several genetic testing companies, including 23andMe, that their genetic tests are considered medical devices and federal approval is required to market them.[14][27] 23andMe first submitted applications for FDA clearance in July and September 2012.[28] On November 22, 2013, after not hearing from 23andMe for six months, the FDA ordered 23andMe to stop marketing its Saliva Collection Kit and Personal Genome Service (PGS) as 23andMe had not demonstrated that they have "analytically or clinically validated the PGS for its intended uses" and the "FDA is concerned about the public health consequences of inaccurate results from the PGS device".[28][29][30] As of December 2, 2013, 23andMe has stopped all advertisements for its PGS test but is still selling the product.[31][32] As of December 5, 2013, 23andMe is only selling raw genetic data and ancestry-related results.[33][34][35]

 

[bauber]

So you're saying I should do a molecular fellowship...

 

[gbwillner]

Personalized medicine IS coming. It just might be a long time before it gets here. It may have a different name when accepted. A revolution IS coming. The sheer amount of data at our disposal now from a single next-gen sequencing test in incomprehensible in comparison to anything that could be done 5 years ago. The rate at which our knowledge of disease-causing and treatment-modifying genetic and epigenetic phenomena is acquired is increasing exponentially.

The idea of the human-created construct of what a disease entails is aged and will see its end. In the future disease will be treated by targeting its underlying mechanism. That is the core of personalized medicine.

Surgical pathologists will always have a role to play in the diagnosis and management of disease. But they can either embrace this reality and "own" the future, or they can marginalize it out of lack of understanding or fear and let Med Onc and Rad Onc and academic genetics departments "own" it and marginalize pathology further and further into obscurity.

 

[pathstudent]

Personalized medicine IS coming. It just might be a long time before it gets here. It may have a different name when accepted. A revolution IS coming. The sheer amount of data at our disposal now from a single next-gen sequencing test in incomprehensible in comparison to anything that could be done 5 years ago. The rate at which our knowledge of disease-causing and treatment-modifying genetic and epigenetic phenomena is acquired is increasing exponentially.

The idea of the human-created construct of what a disease entails is aged and will see its end. In the future disease will be treated by targeting its underlying mechanism. That is the core of personalized medicine.

Surgical pathologists will always have a role to play in the diagnosis and management of disease. But they can either embrace this reality and "own" the future, or they can marginalize it out of lack of understanding or fear and let Med Onc and Rad Onc and academic genetics departments "own" it and marginalize pathology further and further into obscurity.

I completely agree with you but one thing that we need to consider is the cost to society. Let's say sequencing identifies 5 targets in a metastatic pancreatic adenocarcinoma for which there are effective drugs that can block those pathways and extend the patients life for years to indefinitely. The cost of these drugs is for sure going to be in the low six figures per month. We will be spending millions of dollars per year per patient to keep people alive with cancer. I am not sure that is feasible let alone admirable.

 

[LADuck00]

I completely agree with you but one thing that we need to consider is the cost to society. Let's say sequencing identifies 5 targets in a metastatic pancreatic adenocarcinoma for which there are effective drugs that can block those pathways and extend the patients life for years to indefinitely. The cost of these drugs is for sure going to be in the low six figures per month. We will be spending millions of dollars per year per patient to keep people alive with cancer. I am not sure that is feasible let alone admirable.

It may not be feasible but I think it's admirable. Along these lines it will be interesting to see how the new Hep C drugs play out. It'll be $80,000 for a course of treatment that yields an 80% or higher cure rate. The alternative being a liver transplant, and given the length of the transplant list it might as well mean death for a good number of persons (sadly). So we might as well be treating a slowly progressing cancer. MANY people with Hep C are uninsured or under-insured, so footing the bill will be an issue. As many as 2-4 million are chronically infected in the USA alone, so generating statistical significance should not be an issue.

I hear you about costs though. A lot of factors involved but it's certainly a shame.

 

[TheTao]

Personalized medicine is already HERE via transtioanl clinical sciences/translational informatics and major academic institutions, especially those affiliated with major cancer centers. And pathology missed the boat on on leading this field YEARS ago.

 

[pathstudent]

It may not be feasible but I think it's admirable. Along these lines it will be interesting to see how the new Hep C drugs play out. It'll be $80,000 for a course of treatment that yields an 80% or higher cure rate. The alternative being a liver transplant, and given the length of the transplant list it might as well mean death for a good number of persons (sadly). So we might as well be treating a slowly progressing cancer. MANY people with Hep C are uninsured or under-insured, so footing the bill will be an issue. As many as 2-4 million are chronically infected in the USA alone, so generating statistical significance should not be an issue.

I hear you about costs though. A lot of factors involved but it's certainly a shame.

As I have written before. Millions of kids die due to malaria. Most of these deaths could be eliminated with a modest effort by the us and Western Europe. But we are willing to spend trillions so that septangerians and octogenarians can live 3 or 5 more years? The US is a selfish corrupt society and personalized medicine is a prime example of that. Yes it will make corporations trillions and extend the life of the elderly a couple years and it is interesting but it isn't the right thing to so. Healthcare should be about improving wellness and health not just delaying death to the last possible moment.

 

[gbwillner]

As I have written before. Millions of kids die due to malaria. Most of these deaths could be eliminated with a modest effort by the us and Western Europe. But we are willing to spend trillions so that septangerians and octogenarians can live 3 or 5 more years? The US is a selfish corrupt society and personalized medicine is a prime example of that. Yes it will make corporations trillions and extend the life of the elderly a couple years and it is interesting but it isn't the right thing to so. Healthcare should be about improving wellness and health not just delaying death to the last possible moment.

How are the those breadlines, comrade?

Will you be saying these idiocies when/if you develop cancer? I'm sure you'd be happy that the government stopped investing in research that could have helped you so that they could afford to build another billion dollar airplane that can't fly.

 

[pathstudent]

How are the those breadlines, comrade?

Will you be saying these idiocies when/if you develop cancer? I'm sure you'd be happy that the government stopped investing in research that could have helped you so that they could afford to build another billion dollar airplane that can't fly.

Yes. When I have metastatic lung cancer or pancreatic cancer, I hope I have the sense to just go home. It is certainly how I will advise my parents when the time comes. I know that is not what MDA or MSK or DF or my local rad onc/ heme onc wants to hear as that is how they fill their coffers or bank accounts, but I believe it is the most reasonable course of action. Did you see walter whites response to his family about his decision to not pursue treatment during their intervention on breaking bad (yes I know it is a show but it was the best written scene I have ever seen addressing the issue)? I couldn't agree more. But like him I would go for it to, if I needed to cook millions of dollars of meth to provide for my wife and kids.

If it was a 45 year old man or woman, I could understand going an extra mile or two, but ultimately I believe rationing is the only when to keep our do everything system afloat.

The next gen sequencing companies have swept through our community and now every single metastatic tumor is being sent out for sequencing with the hope that some magic bullets is just down the pipe that will the tumor into indefinite remission. Even if this is true, it is not economically sustainable.

If it was as simple for all metastatic solid cancer as it is with cml, where one compound can block the one pathway then maybe it would be doable.

 

[WEBB PINKERTON]

If the current VA system in this country is an omen of what's to come for the rest of us, millions of dollars WILL NOT be spent to keep people alive a few more months.

 

[bberlioz]

In a world with infinite resources, altruism would be ideal and curing malaria would be a priority.
Unfortunately, we live in a world with finite resources and people tend to prioritize their own interests and those of their loved ones.
For those who had to experience the pain of knowing someone with cancer, it is absolutely justified that they donate their assets to a cause that is dear to them. This is their right.

P.S. Pathstudent: you do realize that a large portion of patients have an ECOG score of 0-1? Age should never be a determinant of whether someone is entitled to treatment or not. If they are completely functional, treating their cancer would mean extra years with relatively good quality of life. (I would like to see you declining that!)

 

[deleted314957]

In a world with infinite resources, altruism would be ideal and curing malaria would be a priority.
Unfortunately, we live in a world with finite resources and people tend to prioritize their own interests and those of their loved ones.
For those who had to experience the pain of knowing someone with cancer, it is absolutely justified that they donate their assets to a cause that is dear to them. This is their right.

P.S. Pathstudent: you do realize that a large portion of patients have an ECOG score of 0-1? Age should never be a determinant of whether someone is entitled to treatment or not. If they are completely functional, treating their cancer would mean extra years with relatively good quality of life. (I would like to see you declining that!)

Path student is not talking about people who are completely functional, with a good quality of life who live a couple/few more years.

 

[bberlioz]

Pretty sure he is generalizing here:

As I have written before. Millions of kids die due to malaria. Most of these deaths could be eliminated with a modest effort by the us and Western Europe. But we are willing to spend trillions so that septangerians and octogenarians can live 3 or 5 more years? The US is a selfish corrupt society and personalized medicine is a prime example of that.

What I am trying to say is that it's not about age, but also about functional status. If you are 70-80+ yo and functional, you SHOULD get the care that you deserve regardless of your age. (Heck even if your functionality is limited, you should get whatever treatment is available)

 

[Johnny Sunshine]

I don't think it's a matter of what one deserves, but rather one of simple practicality. Everyone "deserves" the best of everything, regardless of age or function. But that doesn't mean everyone else is obligated to supply them with it irrespective of what it would cost them.

 

[Ziehl-Neelsen]

I was wondering how long it would take for this thread to devolve into the utilitarians versus the objectivists. 30 posts by my count.

Do not then consider life a thing of any value. For look at the immensity of time behind thee, and to the time which is before thee, another boundless space. In this infinity then what is the difference between him who lives three days and him who lives three generations?
--Marcus Aurelius

 

[bberlioz]

Johnny: You're absolutely right! I was arguing that the septangerians and octogenarians with Cancer in America, not only can they afford their care (compared to those in Africa), but they are equally entitled to receive care as those with Malaria in Africa.

Pathstudent was arguing that we should be devoting more resources to Malaria in Africa instead of Cancer in America because:
1) It affects more people on a global scale (Not disagreeing),
2)Malaria tends to affect younger individuals, instead of the older cancer patients. Thus, according to him, instead of "prolonging life in septangerians and octogenarians", we should be devoting more resources to younger people in Africa. (This is an argument that has been used over and over again, which I vehemently disagree)

I was merely providing the counterargument that age should never be part in the equation in determining care.

 

[pathstudent]

In a world with infinite resources, altruism would be ideal and curing malaria would be a priority.
Unfortunately, we live in a world with finite resources and people tend to prioritize their own interests and those of their loved ones.
For those who had to experience the pain of knowing someone with cancer, it is absolutely justified that they donate their assets to a cause that is

Johnny: You're absolutely right! I was arguing that the septangerians and octogenarians with Cancer in America, not only can they afford their care (compared to those in Africa), but they are equally entitled to receive care as those with Malaria in Africa.

Pathstudent was arguing that we should be devoting more resources to Malaria in Africa instead of Cancer in America because:
1) It affects more people on a global scale (Not disagreeing),
2)Malaria tends to affect younger individuals, instead of the older cancer patients. Thus, according to him, instead of "prolonging life in septangerians and octogenarians", we should be devoting more resources to younger people in Africa. (This is an argument that has been used over and over again, which I vehemently disagree)

I was merely providing the counterargument that age should never be part in the equation in determining care.

Random counter points.


I wasn't saying we should spend more on malaria than on cancer.


If what is predicted truly comes to fruition, a world where cancer is kept in check indefinitely by multiple small molecules/targeted therapies, I don't think most american families could afford to pay the millions per year that would cost. Secondly, we are all part of this planet, so what does that say about us that we are willing to devote trillions to let a small percentage of senior citizens, many with a dubious quality of life, live a few extra years while we let millions of young people die which we could prevent with far fewer resources.

Age should absolutely be a factor in allocating care. What are the odds that someone who is 95 lives a mother year versus someone that is 40?

I first started coming to this conclusion back in training. I remember a tumor board conference in hematology where they were talking about how to treat a demented 87 year old, who didn't know who or where he was, for his myeloma because the family wanted "to do everything". I thought someone needs to tell the family to buzz off.

We need to spend our money on keeping people well, not being fat, eating right, screening for diseases when the screening is effective. Not spend money on over treating diseases like prostate cancer and low grade breast cancers, not spending money so a guy with metastatic lung cancer can live to be 72 years and seven months old instead of 72 years and three months old.ô

 

[LADuck00]

The transplant model is certainly one area to observe if we want to study how limited but critical resources are managed. They take a number of patient factors into account. I think most argue the system works as well as you could reasonably expect. Please consider signing your organ donation card and/or becoming a bone marrow donor. A lot of marrow donations can be accomplished with just peripheral blood these days anyway.

 

[WEBB PINKERTON]

3D printers will be making organs someday. Won't need donors. I'm going to print off a brand new Mrs Pinkerton

 

[bberlioz]

Pathstudent: Look, I think you're failing to understand two important points:

1) You keep mentioning the example of " a guy with metastatic lung cancer" who should not receive further treatment because it would only prolong his life and I agree with that. Although this used to be the typical cancer patient in the past, this is NOT your typical cancer patient today.

In the last decade we have gotten much better at detecting and treating cancers at much earlier stages. Whereas in the past we would deal mostly with metastatic/stage IV cancer patients, the majority of oncology patients today are living with stage 1-2-3 with a pretty good quality of life. We are talking about 40-60 year old patients, completely independent,, with families, kids and a job.

Now the big fear in these patients (with treated early stage cancer) is that if the cancer does recur down the line, it often presents with a more aggressive phenotype and is highly likely to be metastatic.

The whole point of research in genomics and personalized medicine is to be able to target this patients with early stage cancer: in order to detect them at an earlier stage, ideally providing some form of adjuvant chemo to decrease risk of recurrence and ultimately increase their long-term survival.

Therefore, despite what you may think, personalized medicine is not so much about prolonging patients' end-of-life, its more about findings ways to intervene at an earlier stage in the disease process in order to increase their long-term survival/quality of life.

2) Again you mention "demented 87 year old". Yes many 80-90 are demented, many of them are frail. Should we be operating on these demented and fail patients and hitting them with radiation and chemo for a relatively low grade prostate CA? Probably not!

But then again, this is not the case of all 80 year olds! As a physician, you should evaluate each patient individually. I've seen some 70-80 year olds who are still independant, living on their own, minimally demented, and relatively healthy (except for some hypertension and diabetes). Are these patients eligible for treatment? Yes they are, and if they have the resources they should.

Stratifying patients by age is very old school. Its very intuitive but its a very flawed system. It is much more adequate to stratify patients based on their performance status in order to determine whether they should get treatment or not. (Which is what most surgeons and clinicians do nowadays)

 

[Muffinho]

Personalized medicine IS coming. It just might be a long time before it gets here. It may have a different name when accepted. A revolution IS coming. The sheer amount of data at our disposal now from a single next-gen sequencing test in incomprehensible in comparison to anything that could be done 5 years ago. The rate at which our knowledge of disease-causing and treatment-modifying genetic and epigenetic phenomena is acquired is increasing exponentially.

The idea of the human-created construct of what a disease entails is aged and will see its end. In the future disease will be treated by targeting its underlying mechanism. That is the core of personalized medicine.

Surgical pathologists will always have a role to play in the diagnosis and management of disease. But they can either embrace this reality and "own" the future, or they can marginalize it out of lack of understanding or fear and let Med Onc and Rad Onc and academic genetics departments "own" it and marginalize pathology further and further into obscurity.

Who is going to be involved in making these new genetic-based treatments? Pharmocogeneticists (all 10 of them)? I'm interested to see how this will be incorporated into medical studies, or if it is something that will be dominated by PhDs.

 

[2121115]

So you're saying I should do a molecular fellowship...

No. But if you don't know your molecular protocols, you will get rolled in the real world. Quick - what is the role of KRAS mutation testing in non-small cell lung carcinoma? If it doesn't naturally roll off of your tongue in the middle of a lunch conversation about kids' soccer games, you need to read up.

 

[BlondeDocteur]

You forgot too about the other side of the life spectrum. What about identifying new Mendelian disorders in syndromic children? I see two clear benefits: 1) giving an official diagnosis gives affected children their required labels to receive services, and the earlier they receive them, the better they do; 2) insight into molecular pathogenesis and mechanism can impact clinical management. For example I just went to a great talk on whole exome sequencing in which kids with profound intractable seizure disorders were sequenced and a new channelopathy was discovered, which impacts anti-epileptic drug choice.

 

[LADuck00]

You forgot too about the other side of the life spectrum. What about identifying new Mendelian disorders in syndromic children? I see two clear benefits: 1) giving an official diagnosis gives affected children their required labels to receive services, and the earlier they receive them, the better they do; 2) insight into molecular pathogenesis and mechanism can impact clinical management. For example I just went to a great talk on whole exome sequencing in which kids with profound intractable seizure disorders were sequenced and a new channelopathy was discovered, which impacts anti-epileptic drug choice.

Nice. Another good example is RET mutations and early (like within the 1st year of life) thyroid removal in these kids. This is a genetic test that is now an (almost absolute) indication for pediatric surgery.

 

[BrainPathology]

Personalized medicine IS coming. It just might be a long time before it gets here. It may have a different name when accepted. A revolution IS coming. The sheer amount of data at our disposal now from a single next-gen sequencing test in incomprehensible in comparison to anything that could be done 5 years ago. The rate at which our knowledge of disease-causing and treatment-modifying genetic and epigenetic phenomena is acquired is increasing exponentially.

The idea of the human-created construct of what a disease entails is aged and will see its end. In the future disease will be treated by targeting its underlying mechanism. That is the core of personalized medicine.

Surgical pathologists will always have a role to play in the diagnosis and management of disease. But they can either embrace this reality and "own" the future, or they can marginalize it out of lack of understanding or fear and let Med Onc and Rad Onc and academic genetics departments "own" it and marginalize pathology further and further into obscurity.

I completely agree with this as well. I'm just uncomfortable with the idea that we aren't doing anything like this now. I think it belittles medicine up to the days before this coming "revolution." There have been other 'revolutions' before that were probably hyped just as much but really just amounted to augmenting what we already did. It's not as if pathologists sat around grunting incoherently while throwing darts at a sheet labeled "benign" and "malignant" before the days of immunohistochemistry (at least not all of them).

We should be able to embrace these new tests (the ones that add information without just adding cost), not because they are completely new and different, but because they are in exactly the same spirit of getting patients the best possible diagnosis for their subsequent treatment using everyt relevant data point available to pathologists making the diagnosis. That's been the goal of pathology forever, it's all just new data to a set of data which is already pretty great relative to the treatments available. (I mean.. it's a close second behind making enough to buy a new Lotus every model year - because let's not kid ourselves about who is always first in the equation here especially on this forum).

 

[LADoc00]

A.) EMBRACE new technologies that have either a proven prognostic or therapeutic role in treatment.

B.) IDENTIFY clear and efficient ways to monetize your role in the process.

C.) PROFIT.

 

[Substance]