Another question on CEA

You can't say "those whom CEA is considered accurate"  - many ppl have normal CEA pre-surgery then develop elevated CEA at recurrence.  Some have elevated CEA pre-surgery then normal CEA at recurrence.   The UK FACS study, followed stage 1,2,3 after surgery with curative intent and recorded CEA readings over 5yrs.   In the followup, ~80% never had an elevated CEA (>5ng/dl) and this 80% had an 11% CRC recurrence rate.   Of the 20%  that had one or more elevated CEA readings, there was 50% recurrence rate.    I've parsed the data and >5ng/dl reading suggests a ~50% chance of recurrence, at >10ng/dl means >80% odds of recurrence., above 37ng/dl this dataset has 100% recurrence.  Cochrane Revieews and the UK national health system report suggest a CEA level of 10ng/dl as threshold for further study (non-smokers only).

The Netherlands "CEAwatch'" study (and several other studies) found that following the RATE of CEA increase is an important criteria, with a threshold ~10% CEA increase per month over 3 months.  The UK national study suggests a RATE of 1ng/dl increase per month increase is the trigger for imaging in the first year, but lower rate triggers later (~0.65ng/dl per month at yr3).

Liver metastases (as Trubrit's comment  suggests) often cause dramatic, exponential increases in CEA - doubling in a few months.  FWIW current thinking is tha CEA actually improves the odds that a liver metastasis will form, and then the high  blood supply in the liver makes release of CEA to the blood more common.

I've never found the stat you are looking for.  CEA readings are just too erratic to determine a clear stating point from which to measure time, and a rapid rise in CEA is likely to be caused by a malignancy, whether it can be imaged or not.   The CEAwatch study (huge national database content) showed the mean time to recurrence ~15-16 months.  Generally they consider tumors found <6months from surgery to be 'synchronous' - tumors they just missed.   LT studies find CRC recurrence from 6months to 115months(~9.5yr) after surgery.  Early recurrence (<24-30 months) are likely to be metastatic liver or lung.  Later recurrence is more likley to be a local recurrence which has somewhat better outcomes.

Perhaps most important, CEAwatch and now several additional recent studies show that extra surveillance (more frequent CEA tests, more CTs&PETs&Ultrasounds, find metastatis at an earlier tumor stage, cause consideraby more surgeries, but don't improve mortality outcomes!!    So detecting recurrence earlier than conventional protocol (typically CEA every 3mo for 2yr, CEA every 6mo for 3yr) isn't helpful or desirable.

When I was diagnosed initially it rose.

Dropped after surgery, then it then increased during chemo (common occurance). 

Dropped after chemo 

Then started rising after met to lung, which was removed. 

Dropped after removal 

Cancer came back in mulitple lymph nodes and multiple mulltiple multiple lung mets (a bunch).    CEA really has not gone up - it is still about 20% less than the reading I had with the one confirmed lung met and has not hit "High" on readings.   Have not bothered testing the last few times in light of the circumtances.  It would show high now, maybe more so with chemo now starting.  Would be interesting to track to see how it goes up and perhaps down as chemo works.  But I will be getting scanned fairly often in light of my condiiton.

Most common infections DO NOT elevate CEA.   Here is a list of some common benign causes of CEA elevation.

• Inflammatory Bowel Disease.
• Pancreatitis.
• Liver disease.
• Tobacco use can lead to elevated CEA levels.
• Diverticulitis.
• Hepatitis.
• Peptic ulcers.
• Hypothyroidism.
• Cirrhosis of the liver.
• COPD.
• Lung infection.
• Pleural effusions.
• Biliary obstruction.
• Treatment with oral 5-FU.
• High serum glutamic-pyruvic transaminase (sGPT, aka ALT) levels.

There are some cases that sugget lithium treatment causes CEA elevation.

--

CEA rising or >10 should trigger further studies.  Some papers suggest that an isolated high measure should be followed up with an additional test (lab error concerns, rare but it happens).

Some tumors of the same cancer produce CEA in the blood stream, some do not.  So CEA is a test with high rates of false positives, and false negatives, therefore poor diagnostic power.

There are some very new circulating cell & circulating DNA tests that still have mediocre true-positive rates (still quite a few false positives), but very low false negative rates, and little or no interference from other benign conditions.

My CEA was high with my initial liver mets. The numbers decreased during chemo at the same rate my tumors shrunk. After my resection, my CEA stayed at 1.  The blood work at my scan that showed the recurrence, my CEA was 1.5. At start of this chemo, 1.6. On the day of my third round, my CEA was 1.2 I realize this is still in the normal range, but I still think the .6 rise is the growing of the small lesions so I am taking it as a sign the chemo is working that the CEA is going down.