DRC- the only problem with extracting data from those 10,000 tests already done is that the negative results aren't going to colonoscopy so you won't know if CG was missing cancers. All you would have are the +'s to compare to colonoscopy.
With a lower prevalence in the 40-50 yr group they would have to do an enormous study to have the power to achieve statistical significance. It might take 30,000 patients to get enough cancers. Not sure they could afford that. Will be interesting to see.
agree 100% - there may be some differences in age groups. My fear is that if USPSTF actually comes out with a lower starting age will Exact have to do a whole DEEP C type study in the 40-50 group or will they allow them to just use old samples?
MKE you asked "Would the false negative rate increase under 50"?
Similar to the Positive Predictive value I discussed before the Negative predictive Value (NPV)(the probability that subjects with a negative screening test truly don't have the disease.) is the true negatives/true negatives + false negatives. I won't go through the calculations but they would be similar at age 40 and 60 because the sensitivity of the test is high and the true negatives are high.
BTW you should always worry about false +'s when the follow up procedure or the treatment is potentially dangerous (this is the whole problem with prostate cancer screening)
DRC I'll try to explain. I don't have time to look up exact numbers but I think the prevalence of CRC in a 60 yr old is about 10 times that of a 40 year old
Positive predictive value(PPV) of a test is the probability that subjects with a positive screening test truly have the disease. (this is very dependent on your pre-test probability of having the disease)
PPV= true positives/true + plus false +
example (Not CRC necessarily):
So lets say only 2/100 40 yr olds have a disease and the false + rate of the test is 13% (like CG may be)
Then the PPV is 2/15 or 13% that a positive test represents a true disease.
But if the disease prevalence at age 60 is 20/100 then 20/20+13=20/33 or 61% chance that the test represents true disease (or 5 times the predictive value of the 40 year old)
This not only adds risk with follow-up testing but drives up cost due to many unnecessary follow-up tests.
You as an individual may be willing to accept this cost and risk but the guideline organizations have to balance the cost/benefit ratio of screening and pick an age where the benefits outway the risk in terms of cost and health in a population and so far they have chosen 50 yrs old.
MKE - Yes the test needs an ordering provider and yes it would be off label under 50. It could still be done.
As a doc I personally would not recommend it for your wife if she is low risk. The benefit of screening tests depends alot on your pre-test probability of having the disease. The younger you are the lower the pre-test probability of having CRC and the greater the likelihood that a +CG is a false positive. So then you are sent for a colonoscopy and the potential complications of that. The doctor would actually be assuming some liability to offer the test off label if a false + result ended in a bad outcome. JMHO. (USPSTF is actually looking at whether to lower the age limit).
Agree. Always best to lower expectations with a new launch. We saw what happened after Q4 with the 25% drop after missing the streets numbers. Would rather have KC controlling the expectations.
They already guided to 10,000. They will not hit 15k this quarter or 30-50 k next quarter. That is wildly optimistic
no I perfectly understood your point - you just forgot to mention all the potential charges when getting a screening colonoscopy
Bio is the below irrelevant regarding CMS coverage of colonoscopy costs? 9/10 CG patients won't have any of these extra costs.
"Colonoscopy: Covered at no cost* at any age (no co-insurance, co-payment, or Part B deductible) when the test is done for screening. If the test results in the biopsy or removal of a growth it is no longer a “screening” test, and you will be charged co-insurance and/or a co-pay (although you still don’t have to pay the deductible).
If you’re getting a screening colonoscopy, be sure to find out how much you will have to pay for the exam. This can help you avoid surprise costs. Patients may still have to pay for the bowel or colon prep kit, anesthesia or sedation, pathology costs, and facility fee. Patients may receive one or more bills for different elements of the procedure from different practices and hospital providers".
Bio - you have very little understanding of clinical medicine - "only the insurer cares about HEDIS, NCQA"- REALLY? Do you know how many Billions of dollars that insurers pay providers in Pay for Performance measures based on HEDIS? Do you know that clinics and hospitals have their HEDIS/NCQA scores open to the public and are very concerned with being in a high performing group? Do you know Hospital system and clinic accreditation is partly based on these HEDIS/NCQA measures. Believe me they are very interested in these measures.
Regarding Kaiser and Geisinger - I'm happy they are successful. But the fact that you can cherry pick a few successful organizations doesn't change the fact that nationally the compliance rate of FIT over a 3 year period is 15% - that is realworld and that is why CG will be successful.
You are not aware of many things. I have used Quest and have had quality initiatives with them. They are used by many insurers as well as other diagnostic companies doing the same thing. They have a success rate of 10-20% as opposed to Exact at 75%. The reason is that FIT is very low margin and they can't devote resources to compliance. Exact can. Have you been to their call center - I have - it's very impressive. When you get $500/test you can devote resources to compliance. Insurers will used it because they can lock in 3 years of a quality measure hit instead of 1 with Fit. FIT cost is not just $23 it is all the cost to the insurers and doctors pay in trying to meet compliance - that is expensive and Exact makes it free to insurers and providers. That is the value proposition.
Biostigg -you are so wrong again. Insurers have been ordering HOME FIT tests for years without the PCP visit to meet HEDIS/NCQA and STARS measures. As soon as USPSTF and NCQA sign off on CG this year insurers will be doing the same thing with CG so they meet their Quality requirements. A Dr visit will NOT be needed. You are only 7 years behind:
Mar 31, 2008
Quest Diagnostics Incorporated , the nation's leading provider of diagnostic testing information and services, today announced the launch of a novel campaign designed to help increase compliance with take-home colorectal screening tests. Titled "It's Time to Challenge Colon Cancer: Do You Have the Guts?," the campaign aims to break down barriers to compliance with fecal occult blood tests and increase screening with these convenient take-home screening tools. The goal of the new campaign is to have five million people complete their annual testing over the next five years.
Agree Quaff - didn't see any info on precancer or stage of cancer. If they are all stage 3 and 4 then not very valuable. Good to see Exact doing work in the plasma area to stay ahead of the curve.
I was thinking - since DEEP C had about 10,000 patients to achieve statistical power, we have already had about 15,000 samples of real world experience. Exact has all of the colonoscopy/pathology data on the CG + patients. We may hear from them that real world CG specificity might be higher than DEEP C. I suspect that when a GI does a Colonoscopy on a CG+ patient they are looking extra carefully for a lesion and will spend more time than they usually would on a low risk patient. If this is the case and specificity is higher than DEEP C that will be a HUGE win and upside surprise for Exact.
Sorry quaffer I totally missed it that these were plasma samples!! I thought it was stool - I'll have to take another look
thanks- interesting abstract. So the markers are fairly specific for organ type (but 83% leaves significant room for false + per DRC's idea). Since this wasn't done specifically with CG markers and their FIT algorithm I'm not sure how this would translate to CG. But KC and his team are pretty smart and I imagine they thought this one through so the algorithms eliminate most if not all of the non colon cancers. But, I'm not on the cover of DNA weekly either :)