H Pylori, PPI use and gastric cancer

​Another article just came out on the relationship between H Pylori infection and gastric cancer, this one in patients who had been successfully treated but are still on PPIs vs H2 receptor antagonsists (see doi:10.1136/gutjnl-2017-314605).

Details:

--Hong Kong study, based on a territory-wide database, of adults who had received outpatient prescriptions for clarithromycin-based triple antibiotic therapy for H Pylori between 2003-2012. They excluded patients who had failed this regimen (needed repeated antibiotic course), and those diagnosed to have gastric cancer (GC) within 12 months after H Pylori therapy, and those with gastric ulcers. they also excluded those with prescriptions for PPIs or H-2 receptor antagonists (H2RAs) within 6 months of GC diagnosis to avoid protopathic bias (prescription of  one of these meds for an early manifestation of gastric cancer)

--PPI users (n=3271) vs non-PPI users (n=60,126): mean age 55, 47% male, follow-up 8 years, 3% smokers, 1% alcohol, 5% GERD/7% dyspepsia, 12% diabetic, 21% htn, 8% hyperlipidemia, 1% obese

--they performed propensity score adjustment (to mathematically equalize the comorbidities between the groups). Potential confounders included sex, smoking status, alcohol, history of gastric or duodenal ulcer, other comorbidities (htn, diabetes, hyperlipidemia, obesity, ischemic heart disease, atrial fibrillation, heart failure, stroke, CKD, cirrhosis) and some meds (statins, metformin, aspirin, NSAIDs, COX-2 inhibitors, clopidogrel)

--H2RA was used as a negative control exposure. They also compared the treated patients to those with H pylori who did not receive treatment

Results:

-- Of 63,397 subjects, 153 (0.24%) developed GC during a median follow-up of 7.6 years (20% of them developed GC of the cardia, 62% in non-cardia regions, rest nonspecified).  All patients with GC were negative for H pylori on biopsy at the time of cancer diagnosis and had underlying chronic gastritis. Median age of cancer patients was 71, and had received H pylori Rx at age 65 (median time from eradication being 4.9 years)

   --PPI use was associated with an increased GC risk, HR 2.44 (1.42-4.20)

   --H2RA was not, HR 0.72 (0.48-1.07).

--the risk increased with duration of PPIs use:

   --≥1 year: HR 5.04 (1.23-20.61)

   --≥2 years: HR 6.65 (1.62-27.26)

   --≥3 years: HR 8.34 (2.02 to 34.41)

--adjusted absolute risk difference for PPIs versus non-PPIs use was 4.29 excess GC (1.25-9.54) per 10,000 person-years.

Commentary:

--Other studies have found that H pylori eradication is associated with a 33-47% decrease in risk of development of GC (see blog below)

--overall, PPIs are pretty safe (and, for better or worse are available OTC without a prescription), though studies have suggested some increase in several diseases as well as significant and perhaps meaningful microbiome changes: see other blogs below

--the combination of PPIs and H pylori infection may lead to increased risk of gastric cancer, with more profound atrophic gastritis (a cancer precursor, especially in those who develop corpus atrophy). PPIs also stimulate gastrin (very high levels), which can lead to hyperplasia of enterochromaffin-like cells. A meta-analysis found the risk of GC is increased 43% among PPI users. 

--it should be noted that the association between H Pylori and GC is a bit more nuanced: there are different strains of H Pylori which affect its virulence and is associated with differing clinical outcomes. one well-studied variant is CagA (cytotoxin-associated gene A), which is associated with higher likelihood of duodenal ulcers (85-100%) as well as with higher likelihood of precancerous lesions and gastric cancer, with a meta-analysis finding a 1.64-fold increased risk over non-cag A H Pylori variants (and this also varies some depending on the specific amino acid sequences in the CagA protein).  In addition, the presence of babA2 (blood-group antigen-binding adhesin) has been reported as having an increased association with duodenal ulcers and gastric cancer.

-- i am not exactly sure how to explain the above association between treated H Pylori and GC  with the subsequent use of PPIs but not H2RAs.  Is there something specific about PPIs in general which interact perhaps with certain variants of H Pylori to create a precancerous state, and continued exposure to PPIs confers a higher risk of GC?? Is it that there is baseline chronic gastritis, perhaps independent of the H Pylori infection (and therefore not changed by treatment of the infection) that itself predisposes to GC, and this gastritis may be severe enough that patients need PPIs vs H2RAs to relieve symptoms??

so, this study brings up a few issues:

-- the bottom line to me, as in prior blogs, is that we should generally use the least strong med which is effective in treating a condition (this time, dyspepsia). we should try to use calcium antacids or H2RAs as first line, increase when necessary to PPIs, and attempt to decrease back to less potent drugs periodically (and certainly after H Pylori, if present, is treated). 

--And, somewhat related, this also brings up to me the concept of routine screening for patients coming from high-risk countries. in my practice, i see lots of patients from Cape Verde, where both H Pylori is pretty endemic and gastric cancer is common (mortality rate of 41.9/100,000: the 2nd highest in the world. see http://global-diseases.healthgrove.com/l/89/Stomach-Cancer ), so i am routinely testing for H Pylori infection and treating. I also have lots of patients from the Dominican Republic where H Pylori is quite common, a recent study finding it present in 62.5%, and the annual incidence of stomach cancer is pretty high at 9.2/100,000/yr vs 3.2/100,000 in the US, though in the US SEER database documented it being 10.1/100,000 in Hispanic persons. The annual incidence of gastric cancer also is quite high in Haiti (7.0/100,000) and Jamaica (11.6/100,000). Guatemala also has a very high mortality from stomach cancer, 6th in the world, at 27.2/100,000 per year​. It may well make sense to test and treat patients coming from endemic H Pylori regions where there is also a high incidence of gastric cancer... 

see: 

http://gmodestmedblogs.blogspot.com/2016/10/h-pylori-regimens-stratified-by.html  for a network meta-analysis of different H pylori antibiotic regimens

http://gmodestmedblogs.blogspot.com/2019/04/h-pylori-eradication-and-reduced-risk.html for the blog on the lower risk of gastric cancer after H pylori treatment

http://gmodestmedblogs.blogspot.com/2015/05/h-pylori-and-nsaids-increased-gi.html  for an article on the increased risk of bleeding peptic ulcers in those with H pylori but also on NSAIDs/low-dose aspirin, the risk being reduced if first treat the H Pylori prior to starting the NSAIDs

http://gmodestmedblogs.blogspot.com/2014/11/gastric-acid-suppression-and-microbiome.html  for one on gastric acid suppression and the microbiome (both gut and lung)

to be added to the email list: contact me at gmodest@uphams.org