Freitag, 19. Oktober 2012

Helicobacter pylori (Part I): should we test-and treat in the very healthy? Points for discussion.


Helicobacter pylori: should we test-and treat in the very healthy? Points for discussion.

Herein I propose that it might be beneficial to treat the ”pathogen” on a case by case basis even though there is no official recommendation for population-wide screening yet. A note of caution: this is based on a preliminary review, but I think it is time to act. As a starter I recommend these two reviews, the authoritative Maastricht IV/ Florence Consensus Report [1a] and a book chapter by Marshall et al. [1b].

H. pylori has been causally linked to atrophic gastritis, functional dyspepsia, peptic ulcer, gastric cancer and some rare conditions e.g. MALT lymphoma, iron loss, etc. The evidence for (other) extra-gastric benefits and harms is weak and inconsistent  [1a, 1b]. The pathogen may promote auto-immune diseases, colorectal adenoma, pancreatic and lung cancer, CVD, disturbed glucose homeostasis, neurodegenerative disorders, and more. Whereas potential benefits from H. pylori infection include: reduced asthma, CVD(!), atopy, inflammatory bowel diseases, weight gain, diarrheal diseases etc. („commensal hypothesis“).

On gastric cancer
Preclincal evidence, observational [1a], and non-randomized studies as well as secondary prevention RCTs (randomised controlled trials)* support a role [4]. As predicted, eradication also improves precancerous lesions, at least somewhat - gastric atrophy may improve but metaplasia does not, Rokkas et al. 2007 [3]: this is the point of no return hypothesis.
Recently, follow-up of a large trial and an updated meta-analysis of primary prevention trials was published and both showed a significant reduction in gastric cancer incidence RR=0.66, 0.46-0.95 with n=4 (see supplementary data, Ma et al. 2012) [2]. The meta-analysis looked at N=4+1+1 studies but even excluding the one secondary prevention (Fukase 2008) and another controversial trial (Leung 2004/Zhou 2008) the result was significant. This was an update of the (then-flawed) 2009 analysis [4b]. However, in this large study there was no decrease in all-cause mortality and a non-significant 30% decrease in gastric cancer mortality. This was mostly expected since the predicted decrease in all-cause mortality was ~ 6% and so the study was underpowered to detect a difference. Nonetheless, this result is decent since their eradication treatment was outdated and reinfection rate must be high. On the other hand, the Asian data is difficult to generalize and the study quality wasn’t perfect.
If we compare H. Pylori eradication to the HPV vaccine, as an example of an approved treatment scheme, it still looks favorable in my opinion: HPV studies were better quality and larger, but the non-cancer benefits of HPV eradication are very modest vs. H. pylori eradication and all data for HPV concerns precancerous lesions only!
All in all, I think the weight of evidence supports population wide screening but the evidence is weak. And if we consider preliminary data on extra gastric effects the case for eradication is stronger, if anything.

But the incidence of gastric cancer is low in the west, particularly in the healthy, isn't it?
Well, yes, but so is the risk of most other diseases, this shouldn’t affect the risk-benefit ratio that much since treatment is very safe.

*to be fair: in this case treatment may only slow cancer progression, if at all, as shown by a recent  study (Maehata Y. 2012)

CVD
H. pylori has been linked to CVD, e.g. stroke in (nested) case-control studies, but prospective cohort studies have been mixed. In fact, the largest study to date found H. pylori to be protective against CVD mortality, but it suffered from multiple testing. (Schöttker et al. 2011) [7]

On esophagal cancer:  This is the best supported problem with H. pylori eradication.
Observational studies suggest a protective role in GERD, barret esophagus and adenocarcinoma but RCTs and meta-analyses do not support a net effect of eradication on GERD, which may even improve afterwards; or worsen in some cases [1a]. Confusingly, some observational studies even showed protection from esophageal cancer depending on the location of infection: “It is associated with a reduced risk of oesophageal adenocarcinoma when it induces  gastric atrophy but an increased risk when it induced a non-atrophic antral-predominant gastritis. It is now also associated with squamous cell carcinoma when it induces atrophic gastritis.” (McColl 2007) [5]
It should be noted that the large experimental study cited above [2] did find a non-significant excess of esophageal cancers.  However, this is not that concerning because GERD can be managed relatively easily and, mechanistically, eradication should only worsen existing GERD.  One should closely monitor for GERD symptoms, nevertheless. Since deaths from e.-adenocarcinoma are about as abundant as deaths from stomach cancer in the US, in the very worst case this might offset gastric cancer benefits.

The conflicting results may be explained by differences in virulence between Asian and non-Asian strain, which also limits our ability to extrapolate from Asian trials.

Other risks of killing H. pylori
Interestingly, most diseases that H. pylori could prevent are relatively easily manageable, early onset and/or low mortality. Whereas the harms of H. pylori are well supported, the benefits are not and they are balanced by other speculative extra-gastric harms, see the reviews.

So Is H. pylori a beneficial commensal?
As far as I know a variation on this theme has been championed by Blaser [7]. However, commensal does not equal beneficial: drift, recent acquisition or antagonistic pleiotropy may explain negative effects of a commensal. "Early benefits, late harms" is emerging  as a plausible mode of action for the pathogen  and Blaser indeed suggests something along this line.
This fact coupled with the problem that gastric atrophy/IM may be irreversible [1a, 3] means that early adulthood is the best time for treatment (this is the second good reason to get rid of H. pylori).

Interactions with dietary salicylates [8a, 8b]
Early reports most probably over estimated salicylate content in food. In a recent systematic review [8b] the mean intake was 3.2/4.4mg mg in a UK population, although, South Indian vegetarian diets contain 12-13 mg. Dietary intakes vary greatly depending on what you consume. Spices and herbs, particularly those consumed in large quantities (say, ~100g), wine, tea, fruit/vegetable, grains and certain legumes are important contributors to one’s intake. Since I cannot access the systematic  review (please, send me the pdf), I am relying on this narrative review as “secondary/tertiary source”. [8a]
Notably, top serum levels of vegetarians can be higher than those of south Indians and both overlap with low dose aspirin users. Top quartile serum levels in a recent adenoma study were also very high. Calculated intakes probably underestimate actual intakes because salicylate precursors e.g. benzoic acid are not included, organically grown food contains more salicylates and they are found in toothpaste and mouthwash (as per [8]). Now consider this: Aspirin – 1mg equals ~0.9mg salicylic acid – intakes as low as 10mg have been linked to GI damage (single RCT), also there is little or only a very flat does response relationship between ~80 and 300 mg of aspirin & GI side-effects (meta-analysis)   –  and to some extent this must extend downwards of 80mg. Obviously, the COX/prostaglandin pathway might also be inhibited and bleeding risk increased by other typically healthy foods and behaviors.
All this taken together could – I speculate – make the “very healthy” exceptionally susceptible to GI damage, enhanced by H. Pylori. As per recommendation, Aspirin (particularly -naïve) users should benefit more from eradication than the general population [1a]:
“H pylori infection is associated with an increased risk of uncomplicated and complicated gastroduodenal ulcers in NSAID and low-dose aspirin (acetosalicylic  acid (ASA)) users.  Evidence level: 2a Grade of recommendation: B
Eradication reduces the risk of complicated and uncomplicated gastroduodenal ulcers  associated with either NSAID or low-dose ASA use.  Evidence level: 1b Grade of recommendation: A
H pylori eradication is beneficial before starting NSAID treatment. It is  mandatory in patients with a peptic ulcer history. Evidence level: 1b Grade of recommendation: A”

I know this is quite speculative, but the evidence fits. (so this is the third good reason to get rid of H. pylori).

Key studies cited only:

[1a] Management of Helicobacter pylori infection—the Maastricht IV/Florence consensus report. Malfertheiner et al.
[1b]Practical Gastroenterology and Hepatology: Esophagus and Stomach.  Chapter 44. Helicobacter pylori. Marshall et al. 2012
[2] Fifteen-Year Effects of Helicobacter pylori, Garlic, and Vitamin Treatments on Gastric Cancer Incidence and Mortality. Ma et al. 2012
[3] The Long-term Impact of Helicobacter pylori Eradication on Gastric Histology: a Systematic Review and Meta-analysis. Rokkas et al. 2007
[4] Clinical prevention of gastric cancer by Helicobacter pylori eradication therapy: a systematic review. Ito et al. 2009
[4b] Meta-analysis: can Helicobacter pylori eradication treatment reduce the risk for gastric cancer?
Fuccio L, Zagari RM, Eusebi LH, Laterza L, Cennamo V, Ceroni L, Grilli D, Bazzoli F.
Ann Intern Med. 2009 Jul 21;151(2):121-8. Erratum in: Ann Intern Med. 2009 Oct 6;151(7):516.
[5]Helicobacter pylori and oesophageal cancer – not always protective. McColl et al. 2007
[6] Atherosclerosis. 2012 Feb;220(2):569-74. Epub 2011 Nov 25.
Helicobacter pylori infection, chronic atrophic gastritis and major cardiovascular events: a population-based cohort study.
Schöttker B, Adamu MA, Weck MN, Müller H, Brenner H.
[7] Nature. 2011 Aug 24;476(7361):393-4. doi: 10.1038/476393a.  Antibiotic overuse: Stop the killing of beneficial bacteria.  Blaser M.
Or a longer review: http://www.ncbi.nlm....10/?tool=pubmed
[8a] Food Funct. 2011 Sep;2(9):515-20. Epub 2011 Aug 30.  Natural salicylates: foods, functions and disease prevention.  Duthie GG, Wood AD.
[8b] A systematic review of salicylates in foods: estimated daily intake of a Scottish population.    Wood A, Baxter G, Thies F, Kyle J, Duthie G.    Mol Nutr Food Res. 2011 May;55 Suppl 1:S7-S14. doi: 10.1002/mnfr.201000408. Epub 2011 Feb 23. Review.

Studies I would like to read but lack access:

Quality of RCTs exploring Helicobacter pylori eradication for the prevention of gastric cancer and preneoplastic lesions.
Sun TT, Wang JL, Fang JY.
Expert Rev Anticancer Ther. 2011 Oct;11(10):1509-19. Review.

Meta-analysis: can Helicobacter pylori eradication treatment reduce the risk for gastric cancer?
Fuccio L, Zagari RM, Eusebi LH, Laterza L, Cennamo V, Ceroni L, Grilli D, Bazzoli F.
Ann Intern Med. 2009 Jul 21;151(2):121-8. Erratum in: Ann Intern Med. 2009 Oct 6;151(7):516.

A systematic review of salicylates in foods: estimated daily intake of a Scottish population.    Wood A, Baxter G, Thies F, Kyle J, Duthie G.    Mol Nutr Food Res. 2011 May;55 Suppl 1:S7-S14. doi: 10.1002/mnfr.201000408. Epub 2011 Feb 23. Review.

1 Kommentar: