OncoCyte - The liquid biopsy

[coroner]

Was listening on the radio the other day to an interview of William Annett, the CEO of this new biotech company that just went public last week. They focus in non-invasive cancer diagnostics using the “liquid biopsy” i.e. blood/urine. They purport this as a means of confirming lung cancer in pts being screened by LDCT w/ inconclusive results instead of getting a bx. via detection of mRNA expression in malignant cells.

Here’s their website and podcast of the CEO’s interview:
www.oncocyte.com
OncoCyte's Annett on Liquid Biopsies to Detect Cancer

I was somewhat annoyed when Annett said the current means of detection of cancer is when a scientist/cytologist looks at the cells and makes a diagnosis. Ummm not quite…more like physicians i.e. pathologists. And, according to him, the “liquid biopsy” will replace the “physical biopsy”. Well, not exactly…tissue, blood, and urine are all “physical”. It’s a difference of invasiveness. Clearly this guy has no medical background, but you would think he should have a slightly better understanding if he was appointed CEO with a relatively modest salary, not to mention a former top dog of Genentech.

Lot of potential pitfalls with this and a lot of the other biotech startups. But a ton of money is being thrown at them. Although, I happen to be invested in one, so I’m only slightly calling the kettle black (but I like to think of it as hedging my bets if technology marginalizes our field). In addition, they have a breast cancer screening test in the pipeline to obviate the need for post-mammogram biopsies of suspicious lesions. And, another one is in the works for bladder cancer. I’m not worried that this will be cutting into our volume anytime soon, but curious what others think. Are companies like this the wave of the future or do we have another Theranos on our hands and OncoCyte will end up being on the list of soon-to-be casualties when the biotech bubble bursts?

---

Members don't see this ad.

 

[Sulfinator]

Was listening on the radio the other day to an interview of William Annett, the CEO of this new biotech company that just went public last week. They focus in non-invasive cancer diagnostics using the “liquid biopsy” i.e. blood/urine. They purport this as a means of confirming lung cancer in pts being screened by LDCT w/ inconclusive results instead of getting a bx. via detection of mRNA expression in malignant cells.

Are companies like this the wave of the future or do we have another Theranos on our hands and OncoCyte will end up being on the list of soon-to-be casualties when the biotech bubble bursts? .

I predict that this company's strategy will replace lung biopsy as a definitive approach to a lung cancer diagnosis about as much as a PSA test does for a prostate cancer diagnosis.

 

[Over9000]

Guy sounds like a handshaking douchenozzle. The "scientist makes diagnosis" bollocks that is circulated around about pathologists infuriates me.

 

[Ziehl-Neelsen]

The "liquid biopsy" may not replace the biopsy for initial diagnosis of cancer, but IMO cell free DNA analysis on peripheral blood samples will replace a significant number of biopsies of suspicious lesions identified on scans for recurrence/metastasis within the next 10 years. I see the analysis of cfDNA significantly encroaching on the realm of traditional histologic cancer diagnosis. I'm not sure if the radiologists care, but I foresee that this technology will significantly reduce the number of scans for recurrent/metastatic disease. For those starting pathology residency now and planning for a 30 year career in primarily AP, all bets are off.


http://www.ncbi.nlm.nih.gov/pubmed/24553385
http://jama.jamanetwork.com/article.aspx?articleid=2389341#jpc150004r23
http://www.ncbi.nlm.nih.gov/pubmed/25579085

 

[pathstudent]

Agree completely with the above.

If you are an alpha type medical student with broad talents and interest I strongly suggest you pursue subspecialty surgery. Regardless of the model of medical reimbursement you will be the best off.

If you were the introverted, misanthropic, cynical medical student who hated all his third year rotations, than you should probably still pursue pathology as there is no other option for you.

 

[Alteran]

There are plenty of disruptive diagnostic technologies coming down the pipeline. I'm personally dealing with some of them in my routine practice. This particular one has been mentioned (or something super similar to it) in our organization. It is my understanding that this kind of technology is only for use in patients for whom it is physically not possible to obtain tissue for molecular testing. A bunch of oncologists have gotten their collective hands slapped by my hospital system for ordering this out of convenience and were told that further ordering of this test outside of this very specific indication is grounds for disciplinary action. As it is, insurance considers this experimental and won't pay a dime for it, as far as I've been told.

 

[yaah]

It's a fascinating technology but the actual clinical implications are a bit far off and unclear I would think. This could replace a lot of biopsies. If you have an obvious lung cancer then this could obviate need for biopsy because you know there's a tumor that's not surgically curable and if you find the right molecular abnormalities you can treat it. But the supposition that most cancers secrete detectable specific abnormalities seems questionable to me. I doubt most GU, endometrial, or oral cavity cancers do, for example.

The biggest problem is when you use it. Do you wait until you have a mass? Do you use it for screening? If you detect something, what do you do about it? What if you find something that is often seen in pancreatic cancer but there is no pancreatic mass? Do you do a Whipple? Or do you wait for it to show up, at which point the testing hasn't really given you much of anything anyway. Are we going to advocate chemotherapy for a molecularly-detected abnormality? How specific is it actually?

These are huge questions which will require a lot of independent study (not just their studies which will be biased to show the outcome they desire).

I really really hope that the significant questions and criticisms raised about this technology are not dismissed as being raised out of fear or traditionalism, but I worry that is going to happen - you can already see it. A lot of the commentary on this focuses on comparison to "old fashioned methods" or "replacing inefficient opinion-based diagnosis with facts" and such. No doubt this will provide a great leap in cancer care and detection, but we have to be responsible about it and ask the right questions.

 

[Ziehl-Neelsen]

Perfect example of how I foresee cfDNA changing surgical pathology from this past week...

I signed out the biopsy and hemicolectomy specimens of a patient with colon adenocarcinoma.

Tumor was just barely T3. Margins negative. 21 nodes negative (I grossed it myself, and I can "mash fat" with the best of them) and no tumor deposits. Carcinoma was high grade and focal perineural invasion was identified. MMR IHC showed no loss of MMR proteins.

Well now the oncologist is in a pickle. IIA disease where adjuvant FOLFIRI has shown a 2% advantage in recurrence free survival at 5 years and no clear benefit for OS, except in patients with "high-risk" features (of which this patient has 2--PNI and high grade histology). What to do?

Well obviously what affects RFS and OS is the presence of undetectable metastatic disease present at the time of surgery--virtually every pathologic parameter that we seek on a colectomy specimen (lymph node metastasis, tumor deposits, PNI, LVSI, perforation, etc.) is geared toward predicting whether the "horse is already out of the barn" and whether the benefits of adjuvant systemic chemotherapy outweigh the risks. Well what if the patient had NGS performed on the initial tumor and cfDNA analysis (after an appropriate interval) could be used to detect DNA with that specific molecular signature in the patient's blood stream? Would that affect the oncologist's decision regarding adjuvant chemotherapy?

Hell, one could argue (and I do) that the detection of cfDNA that matched the carcinoma in the patient's blood would abrogate every single pathologic feature in the tumor that I spent 4 years learning how to identify and report. Who cares if the nodes are histologically negative when the patient has circulating tumor DNA 4 weeks after surgery?

I can even contemplate a day when the colectomy specimen never even goes to the pathology lab. From an oncology standpoint, the only thing that matters after colectomy is whether the surgery "cured" the cancer. Who cares what a pathologist has to say if cfDNA can reliably determine whether there is residual disease?

In the scenario I describe, the only thing I'm needed for is the confirmation of cancer on the initial biopsy that justifies surgery. I don't believe that diagnostic histology will completely go away in the next 30 years, but the volume of specimens (IMO) will drastically decrease. And lest anyone call me a "chicken little" hypocrite, I put my money where my mouth is. I initiated large positions in ILMN (and other companies I consider to be at the vanguard of this technology) during residency when I started researching circulating tumor cells, cell free DNA, and next generation sequencing, and I have been methodically moving over to the CP side of the house since I graduated residency. My momma didn't raise no fool.

#WinterIsComing

 

[forhumanity]

Well if the lymph nodes are positive, you can save money on cDNA? How specific is the cDNA for said cancer, and how is it any different from CEA (for colon), PSA (for prostate) and CA-125 (for ovarian)?

 

[pathstudent]

Supposedly if your amplify anyone's blood long enough you can detect cells that have the philly chromosome. By your reasoning we would all be on Gleevec. But obviously the immune system takes these cells out for almost all of us.

I don't think detecting tumor DNA would necessarily mean you are going to likely die of metastases without treatment. Moreover, chemotherapy doesn't really cure most metastatic cancers.

Perfect example of how I foresee cfDNA changing surgical pathology from this past week...

I signed out the biopsy and hemicolectomy specimens of a patient with colon adenocarcinoma.

Tumor was just barely T3. Margins negative. 21 nodes negative (I grossed it myself, and I can "mash fat" with the best of them) and no tumor deposits. Carcinoma was high grade and focal perineural invasion was identified. MMR IHC showed no loss of MMR proteins.

Well now the oncologist is in a pickle. IIA disease where adjuvant FOLFIRI has shown a 2% advantage in recurrence free survival at 5 years and no clear benefit for OS, except in patients with "high-risk" features (of which this patient has 2--PNI and high grade histology). What to do?

Well obviously what affects RFS and OS is the presence of undetectable metastatic disease present at the time of surgery--virtually every pathologic parameter that we seek on a colectomy specimen (lymph node metastasis, tumor deposits, PNI, LVSI, perforation, etc.) is geared toward predicting whether the "horse is already out of the barn" and whether the benefits of adjuvant systemic chemotherapy outweigh the risks. Well what if the patient had NGS performed on the initial tumor and cfDNA analysis (after an appropriate interval) could be used to detect DNA with that specific molecular signature in the patient's blood stream? Would that affect the oncologist's decision regarding adjuvant chemotherapy?

Hell, one could argue (and I do) that the detection of cfDNA that matched the carcinoma in the patient's blood would abrogate every single pathologic feature in the tumor that I spent 4 years learning how to identify and report. Who cares if the nodes are histologically negative when the patient has circulating tumor DNA 4 weeks after surgery?

I can even contemplate a day when the colectomy specimen never even goes to the pathology lab. From an oncology standpoint, the only thing that matters after colectomy is whether the surgery "cured" the cancer. Who cares what a pathologist has to say if cfDNA can reliably determine whether there is residual disease?

In the scenario I describe, the only thing I'm needed for is the confirmation of cancer on the initial biopsy that justifies surgery. I don't believe that diagnostic histology will completely go away in the next 30 years, but the volume of specimens (IMO) will drastically decrease. And lest anyone call me a "chicken little" hypocrite, I put my money where my mouth is. I initiated large positions in ILMN (and other companies I consider to be at the vanguard of this technology) during residency when I started researching circulating tumor cells, cell free DNA, and next generation sequencing, and I have been methodically moving over to the CP side of the house since I graduated residency. My momma didn't raise no fool.

#WinterIsComing

 

[pathmonkey]

Perfect example of how I foresee cfDNA changing surgical pathology from this past week...
Hell, one could argue (and I do) that the detection of cfDNA that matched the carcinoma in the patient's blood would abrogate every single pathologic feature in the tumor that I spent 4 years learning how to identify and report. Who cares if the nodes are histologically negative when the patient has circulating tumor DNA 4 weeks after surgery?

The reason you care about node status is because its still probably an independent predictor of prognosis, and it gives near instant feedback. If you're going to histologically examine the tumor, ALSO looking at nodes gives more information with negligible marginal cost.

Until you know exactly what the clinical implications of your + cfDNA are. . .and the only way you can for sure is to look at a long term case series of patients with cancer you've tested for this. . .I don't think you can assume its going to entirely replace old fashioned light microscopy. Maybe . . .maybe not.

I can even contemplate a day when the colectomy specimen never even goes to the pathology lab. From an oncology standpoint, the only thing that matters after colectomy is whether the surgery "cured" the cancer. Who cares what a pathologist has to say if cfDNA can reliably determine whether there is residual disease?

"If". Maybe you're right that this is possible, but right now its science-fiction.

Remember, examination of a resected colon can tell you a lot more than just whether or not it has cancer. Quality control for surgeons is one use. Even if your cDNA test can tell you if there is residual disease, doesn't it still "matter" if that residual disease is still there because the pt had metastatic disease at the time of surgery, or because the surgeon did an inadequate resection?

Do you have a cfDNA test for diverticulosis? Amebiasis? Scleroderma? Lymphangiomas? Etc. Yes, I realize molecular tests for many of these things are possible, but simple H&E examination can "test" for a massive number of different things (neoplastic AND non-neoplastic) all at once in one quick, cheap exam, more than any realistically likely molecular panel can.

In the scenario I describe, the only thing I'm needed for is the confirmation of cancer on the initial biopsy that justifies surgery.

Why would you even be needed for that?

**IF** you have accurate, cheap, reliable molecular testing for metastatic disease, then you have it for primary disease too, right? So if your imaging analysis shows something that looks like a colorectal carcinoma, you throw the biopsy in the PCR machine, and you might have a "cancer" or "not cancer" answer faster than you could even generate a slide, plus now you have some sort of baseline DNA "footprint" you can use to compare against to later help identify metastatic disease.

I don't believe that diagnostic histology will completely go away in the next 30 years, but the volume of specimens (IMO) will drastically decrease.

I think there are reasons why H&E histologic exam has been the standard for over 100 years, and I think predicting its demise is premature. Technologic ability to image lesions and sample them, is only getting better; we're getting KINDS of samples today that we didn't even ten years ago. Although labor-intensive, relative to most molecular testing, ordinary diagnostic histology is still actually cheap AND fast. Yes, I realize options are increasing, and costs decreasing, but what cancer-diagnostic molecular test can you get done within one day for $75?

Of course not all histologic testing is about making CANCER diagnoses. Are you going to do blood-based molecular testing for esophagitis or contact dermatitis? Are you going to use it to stage liver fibrosis? Lets say you have a skin tumor. Which is going to be more clinically useful and cost effective. . .simple H&E with margin evaluation , or some necessary PANEL of molecular tests (to look at over a dozen possible types of skin cancer) without?

And lest anyone call me a "chicken little" hypocrite, I put my money where my mouth is. I initiated large positions in ILMN (and other companies I consider to be at the vanguard of this technology) during residency when I started researching circulating tumor cells, cell free DNA, and next generation sequencing, and I have been methodically moving over to the CP side of the house since I graduated residency. My momma didn't raise no fool. #WinterIsComing

Historically, biotech has been high-risk. . .most "miracle" new technologies don't pan out. But we'll see.

 

[pathmonkey]

The "liquid biopsy" may not replace the biopsy for initial diagnosis of cancer, but IMO cell free DNA analysis on peripheral blood samples will replace a significant number of biopsies of suspicious lesions identified on scans for recurrence/metastasis within the next 10 years. I see the analysis of cfDNA significantly encroaching on the realm of traditional histologic cancer diagnosis. I'm not sure if the radiologists care, but I foresee that this technology will significantly reduce the number of scans for recurrent/metastatic disease. For those starting pathology residency now and planning for a 30 year career in primarily AP, all bets are off.

On the last thing, again, wake me up when your blood test can moot intraoperative margins on a squamous cell from the face.

On radiology, I could see good blood-based molecular testing having a role in screening post-resection cancer patients in remission. But it probably wouldn't apply to certain tumor types, and it still wouldn't entirely eliminate imaging.

If you're screening cancer patients for recurrence/residual disease with (say) annual blood based molecular testing, and your screen comes up positive, what do you do next?

Assuming you actually want to treat this, you now have to figure out if the patient has metastatic disease, locally recurrent disease, both, or neither (perhaps a second primary).

This matters; if they only have locally recurrent disease, that could be amenable to further surgery or radiotherapy and you need to know exactly where it is to plan your field. If they have metastatic disease, maybe you want chemotherapy. Etc.

How are you going to make this distinction with the patients blood? You can't. . .you need to image the patient. So much for no more imaging.

OK, lets say you image, but can't find a definite tumor. IE you've molecularly detected residual disease that is physically too small to be identified reliably by imaging. Now what?

PET Scan? Wait. . .image again? How often? You're basically just kicked the patient back into a routine imaging screening algorithm.

Now lets say your imaging finds a possible small lesion, but the radiologist isn't entirely sure what it is. Now you need to biopsy it. Since you already know the patient has tumor DNA floating around in their bloodstream, doing your DNA test on this biopsy may not be reliable; it may show up as positive regardless of whether or not its "cancer".

If only there were some way for someone to look at a tissue sample and determine if there were cancer there.

 

[coroner]

If you're screening cancer patients for recurrence/residual disease with (say) annual blood based molecular testing, and your screen comes up positive, what do you do next?

Assuming you actually want to treat this, you now have to figure out if the patient has metastatic disease, locally recurrent disease, both, or neither (perhaps a second primary).

This matters; if they only have locally recurrent disease, that could be amenable to further surgery or radiotherapy and you need to know exactly where it is to plan your field. If they have metastatic disease, maybe you want chemotherapy. Etc.

How are you going to make this distinction with the patients blood? You can't. . .you need to image the patient. So much for no more imaging.

The end-all-be-all isn’t necessarily to treat or not to treat. But, other adjunct benefits e.g. surveillance, reducing M&M, earlier detection, etc. All factors that can contribute to coming up with a better algorithm as you put it, provided such molecular tests are proven to be clinically significant. To paraphrase the above series of proposed questions with a terse answer, it depends on the diagnosis, stage, context, nature of disease, etc.

For example, surveillance in bladder (urothelial) cancer is important because it has a 50-80% recurrence rate. Do you remember Urovysion FISH? Well, it has a 75-90% sensitivity rate to detect recurrence which is very similar to cystoscopy (although CIS is not detected 22% of the time by cystoscopy) . Regardless of some urologists advocating for cystoscopy vs Urovysion for surveillance because it’s current standard diagnostic procedure (not to mention they can charge for it), any patient would opt for a non-invasive molecular test such as this with similar sensitivity vs. a cystoscopy that comes with the discomfort and higher costs which makes the decision even easier. If such a molecular test is positive, then a follow-up cystoscopy with biopsy can be performed rather than going thru cystoscopy q3mo. x 3 yrs. And no, I’m not a shareholder/spokesperson for Abbott Labs.

I predict after that tide washes out, the will be a select few of these molecular diagnostic tests to have proven diagnostic and therapeutic implications. The ones that gain acceptance from the general medical community because of their scientific validity, cost effectiveness, and (most importantly) approved reimbursement from CMS will be the ones to survive. All the speculative questions about if and when to order and what to do with the results will be hashed out in some future set of NCCN guidelines. Future screening+imaging+ biopsy algorithms will incorporate +/- ordering cfDNA, mRNA, MSA, FISH, etc. somewhere. Hence, their role will be an “ancillary” test in adjunct to tissue-based diagnosis rather than substitute. By the time it gets to the Star Trek phase of a blood test telling it all (Theranos 2.0 anyone?) we’ll all be gone…

 

[yaah]

The big questions will probably need extensive, non-biased and long-term studies.

You need to compare to existing practice and disease course to see if it makes any difference.

I don't know how this will shake out but a lot of the newer molecular tests and advances haven't made much clinical difference except for the ones that identify a mutation that induces sensitivity to a specific drug, or can refine prognosis. A lot of them just show all these abnormalities, but no one has any real idea what to do about them.

 

[coroner]

A new “old” player is trying to crack the liquid biopsy market it seems. Two MGH-based researchers have gotten $30 million in funding from Johnson& Johnson to develop their technology. Dubbed the “iChip”, it captures whole/circulating tumor cells (CTCs) by filtering blood first and then looking for tumor cells left behind.

Inside the race to diagnose cancer from a simple blood draw

The above article is titled as such because its referring to the current market as a so-called “race” because: 38 companies in the U.S. are out there developing liquid biopsy technology. As recently as 10 yrs. ago, there was only 1. OncoCyte was just one of those that I heard about earlier this year.

The market is projected to become a $29 billion industry according to this article. It’s also currently drawing in hundreds of millions of dollars and big-time investors such as Bill Gates and Jeff Bezos.

It’s worth mentioning that some companies have also folded e.g. ‘On-Q-Ity’ was a Waltham, MA based liquid biopsy startup that folded in 2013 because their product wasn’t going to roll out for 5-10 yrs. which was taking too long. The i-Chip founders have apparently been on a ten year journey; but, their approach seems unique plus they recently secured funding which should keep them afloat for enough time if their product does take off…

 

[Deucedano]

All of these liquid biopsy technologies sound amazing when they are written in these articles and when you hear the pitches from proponents of said technology, but when you look at the actual data it is very different. The molecular attendings at my institution are more than skeptical regarding the clinical utility of these technologies and suggest companies may overstate the senitivity of their assays. Some of the tests Guardant Health offers have proven useful, but only in situations when there is a known cancer. What happens when you start screening normal people and you find "tumor cells" in their blood without a mass or clinically evident disease. The cutoffs for allele frequency of calling a mutation is random and varies from company to company and institution to institution. Some use allele frequencies as low as 0.1% which can be found in a large proportion of the population. Does that mean we should start screening those people for cancer development? Where do we start? Will it be cost effective after a full body CT and PET scans? Even if you find a handful of mutations we still dont know what they mean.

Many of these companies promote how they can detect cancer sooner, but they have no idea what these mutations actually mean. We barely understand baseline mutations in otherwise healthy people. Having said all of that, I personally believe, in time, that these types of technologies will replace a proportion of what anatomic pathologists do, but much much more research needs to be done in order to make sense of it all especially in the era of cost effective medicine. These technologies need to prove their worth in order to be deemed worthy.

 

[yaah]

First off, I hate the two terms most commonly used in discussion of these assays
1) "Liquid biopsy" - No. It's a complicated molecular assay performed on the blood. Is a urine cytology a "liquid biopsy"? How about an FNA? Hate this term.
2) "Simple blood test" While yes, it is simple to collect, nothing else about the test is simple. The interpretation, the performance, and the clinical implications. I understand doing this test on patients with known malignancy, looking for early recurrence. Or using it to detect treatable mutations in known cancers. But what about someone with a lung mass who gets a positive test? Do you treat it without confirming it's actually cancer? The biggest question is how common these abnormalities are in patients without disease. Thus, no doubt some patients will start getting them as "screening" tests or as an evaluation if they have a vague imaging abnormality. What do you do if it's positive? What do you do if it's negative?

There are so many questions that need answering before these tests should become commonly utilized. Like I said, the concept of using it to follow known cancer patients is probably the best option and will provide likely significant clinical benefits in some patients. But how many patients currently in remission will have detectable abnormalities that don't progress to anything clinically significant? I just worry about the slippery slope of what happens.

 

[yaah]

Many of these companies promote how they can detect cancer sooner, but they have no idea what these mutations actually mean. We barely understand baseline mutations in otherwise healthy people. Having said all of that, I personally believe, in time, that these types of technologies will replace a proportion of what anatomic pathologists do, but much much more research needs to be done in order to make sense of it all especially in the era of cost effective medicine. These technologies need to prove their worth in order to be deemed worthy.

Exactly. What happens if someone at risk for lung cancer gets the test and it's positive, and imaging is clear except for pre-existing 2 mm nodules? Closer surveillance? Biopsy? More procedures? What happens if a person with a lung nodule gets the test and it's positive, so they take out the largest nodule and it's a granuloma?

 

[WEBB PINKERTON]

I still can't over how many women are getting a biopsy based solely off an HPV result. We have been seeing this frequently. Normal paps throughout their lives but due to the new guidelines are getting an HPV with pap and testing positive. Yet to see one of these biopsies show a significant lesion. Nothing surprises me anymore. Seems like every "improvement" we make, we end up with more costs and invasive procedure.

 

[coroner]

What happens when you start screening normal people and you find "tumor cells" in their blood without a mass or clinically evident disease.

Well, you don’t screen normal people without sxs. Just like normal, healthy people don’t get full body body CTs looking for possible masses. The primary application so far is surveillance in pts. w/ a previous dx of ca., earlier detection in high risk pts., reducing M&M in pts. with proven ca., or poor candidates for invasive procedures.

What happens if a person with a lung nodule gets the test and it's positive, so they take out the largest nodule and it's a granuloma?

False positives happen all the time in medicine. Part of implementing this technology is to hopefully provide something that is close enough in sensitivity to provide an adjunct to what we already have available.

I could just as easily ask what happens when someone has a mass with microcalcs detected on mammography and on subsequent biopsy it turns out to be a sclerosed fibroadenoma with stromal microcalcifications.

I’m not an expert by any means, but the basis of this technology is that the specific mRNA detected in lung ca. isn’t amplified by granulomas.

Having said all of that, I personally believe, in time, that these types of technologies will replace a proportion of what anatomic pathologists do, but much much more research needs to be done in order to make sense of it all

There are so many questions that need answering before these tests should become commonly utilized. Like I said, the concept...will provide likely significant clinical benefits in some patients.

I do agree that long term scientific validity is needed for effective implementation to benefit the pt. And there's a lot of players on the market.
Hence, why I sort of think this is the bubble of a burgeoning technology. Only a few will survive that have proven diagnostic & therapeutic implications. With this will come algorithms of screening+imaging+ biopsy +/- new molecular testing. It may take awhile for this to hash out though…

I still can't over how many women are getting a biopsy based solely off an HPV result. We have been seeing this frequently. Normal paps throughout their lives but due to the new guidelines are getting an HPV with pap and testing positive.

The HPV test is only recommended for primary screening (not routine use or follow up surveillance). It has been shown that in cytology negative women, about 20% who test positive for HPV 16 or 18 will develop HSIL.

"The Elevated 10- year Risk of Cervical Precancer and Cancer in Women w/ HPV 16 & 18..." Journal of the National Cancer Institute, July 2005

At this point, these are still recommendations, not official guidelines. But, it may come to that. However, with the FDA approval of the Roche-Cobas test in Spring '14, this will likely lead to fewer Paps.

 

[WEBB PINKERTON]

Paps have been on a downward trajectory for a while now. Our lab and the many others in the area have easily seen a 50 percent drop in recent years with a huge rise in HPV testing. I am seeing a lot of biopsies on women aged 30+ who test positive for any type of HPV (not just 16/18), who have a history of negative paps. I don't know if the physician or patient is pushing for the agressive follow-up. Luckily none have shown any significant lesions (knock on wood).

If I were investing in circulating tumor cell companies, I wouldn't plan on it paying out anytime soon. If you want to invest money, get into rental property. A nice income to have coming in with the instability of the pathology job market.

 

[Deucedano]

Well, you don’t screen normal people without sxs. Just like normal, healthy people don’t get full body body CTs looking for possible masses. The primary application so far is surveillance in pts. w/ a previous dx of ca., earlier detection in high risk pts., reducing M&M in pts. with proven ca., or poor candidates for invasive procedures.

Thats what the article you posted is saying. I agree it is worth it in symptomatic patients, but at that time wouldnt it already be too late. How would that be better than imaging and biopsy? You still cant make a diagnosis based solely on molecular methods and you would have to image and people in order to get a diagnosis. Given how little we know at this point that would result in a huge increase in imaging. The biggest benefit of this technology is screening normal "healthy" people and detecting cancer before a clinically evident mass develops or monitoring disease response. I believe this is the future of medicine, but we are decades away from this.

False positives happen all the time in medicine. Part of implementing this technology is to hopefully provide something that is close enough in sensitivity to provide an adjunct to what we already have available. I could just as easily ask what happens when someone has a mass with microcalcs detected on mammography and on subsequent biopsy it turns out to be a sclerosed fibroadenoma with stromal microcalcifications. I’m not an expert by any means, but the basis of this technology is that the specific mRNA detected in lung ca. isn’t amplified by granulomas.

No mRNA is specific to any one disease and there lies the problem. You can find low level mutations in normal "healthy" people that only increase with age. These tests only give you a probability based on the profile of mutations, which could be useful if used in the appropriate clinical setting. However, we all know oncologists will order these tests on everyone. Then you get stuck with ambiguous results like Yaah pointed out and could end up doing more unnecessary procedures. I agree with you Coroner that this technology is the future, but its moving too fast into the clinical space without validation. Biotechnology companies are taking advantage of the national movement towards "precision" medicine by offering these tests touting they are specific and more objective, but have not truly validated their claims nor made an attempt to link their results with clinical data.

 

[yaah]

Well, you don’t screen normal people without sxs. Just like normal, healthy people don’t get full body body CTs looking for possible masses. The primary application so far is surveillance in pts. w/ a previous dx of ca., earlier detection in high risk pts., reducing M&M in pts. with proven ca., or poor candidates for invasive procedures.

But it's a "Simple blood test"!!!

We do screen normal people without symptoms in many ways. Not CT scans right now but many others are like that.

You don't think there will be a sizeable population of people (and maybe even some Drs) who will advocate getting this type of test for themselves even if they are healthy? I do.

Insurances will have to decide what to pay for though. They may not pay, at least initially, for the test in healthy people. But what if there's a positive test? Does insurance then pay for the workup to find the tumor?