COMMENTARY ON HOT TOPICS
We read with great interest the recent article by Dinis-Ribeiro et al reporting the first European Guidelines on management of precancerous conditions and lesions in the stomach (MAPS), and strongly recommend it to readers. Despite gastric cancer incidence is decreasing, such a neoplasia remains the fourth most prevalent tumor and second most frequent cause of cancer-related mortality in the world. The endoscopic-based screening programs performed in a few Asian countries, where the gastric cancer incidence is extremely high, are not feasible in other countries due to a distinctly lower frequency of such a neoplasia. Therefore, in the Western countries, surveillance of precancerous conditions [gastric atrophy and intestinal metaplasia (IM)] and lesions (dysplasia) is the only reliable procedure able either to reduce gastric cancer onset - i.e., by removing dysplasia areas at endoscopy or to diagnose an already-developed cancer in an early stage, so that patient survival is distinctly improved[4,5]. However, the American Society for Gastrointestinal Endoscopy Guideline recommended against the surveillance for patients with IM. Until few months ago, no European guidelines were available on the management of these precancerous lesions, leaving both gastroenterologists and general practitioners (GP) to empirically manage these patients without any actual reference standard. For this reason, Dinis-Ribeiro et al should be commended for having organized a workshop involving a vast panel of experts, in order to deliver the first European guideline on such a topic.
According to these new guidelines, it has been recommended that patients with extensive atrophy gastritis (AG) and/or extensive IM - i.e., involving both antral and gastric body mucosa - should be offered endoscopic surveillance (evidence level 2++, recommendation grade B) every 3 years (evidence level 4, recommendation grade D). On the contrary, no scheduled endoscopic/histological control has been advised for those patients with precancerous conditions confined to the antrum. Therefore, both gastroenterologists and GPs have now a “road map’’ to systematically schedule the surveillance in patients with either AG or IM.
Although any guideline is better than no guideline, we would further discuss some potential flaws entailed in the management recently recommended. First, it might appear questionable that the same endoscopic follow-up has been advised for both AG and IM patients. In a nationwide study, the 10-year gastric cancer incidence was estimated to be 0.8% and 1.8% in 22 365 AG and 61 707 IM patients, respectively, corresponding to an adjusted yearly incidence of 0.055% and 0.1%. Given the two-fold different gastric cancer risk, perhaps a differently scheduled endoscopic surveillance should be proposed for AG and IM patients. Differently from IM, a strict follow-up in AG patients may be not cost-effective in Western countries.
Secondly, appropriateness of the suggested interval for endoscopic follow-up is not well documented. The 3-year interval selected for patients with extensive AG or IM does not appear to be corroborated by any prospective study. Indeed, the panel of experts downgraded this statement as level 4, grade D. However, at least two studies demonstrated that only 36% and 38% of detected gastric cancers were in an early stage (i.e., stage I disease), when scheduling the endoscopic surveillance interval at 1- or 2-years, respectively[11,12]. Therefore, despite patients underwent a more intensive follow-up than the 3-year interval now officially recommended, gastric cancer was diagnosed in an advanced stage in as many as 62%-64% of the cases. Based on these observations, an even worse scenario cannot be excluded when a 3-year interval follow-up is to be implemented in clinical practice. The dismal prognosis of gastric cancer diagnosed in an advanced stage poses ethical concerns about recommending a 3-year surveillance interval for IM patients. On the other hand, it is also unquestionable that an appropriate use of endoscopic procedures is essential to the rational use of finite resources. To dissipate economic resources in performing yearly endoscopic controls in all IM patients - most of which would never develop gastric cancer - would also be unethical. A possible solution could be represented by a patient-tailored approach. Similarly to the extensive spreading of AG or IM in the stomach - the only risk factor considered in the MAPS guidelines - several studies demonstrated that other factors increased gastric cancer risk, including IM type, first-degree family history of gastric cancer, and smoking habit. Presence of incomplete type IM significantly increased the hazard ratio of gastric cancer as compared to complete IM (hazard ratio: 11.3, 95%CI: 3.8-33.9). A first-degree family history of gastric cancer also increases such risk by 2.6-3.5 times, with a calculated attributable risk of 8%[14,15]. Indeed, the gastric carcinogenetic cascade in these subjects seems to start earlier than in controls. A meta-analysis, considering 14 442 cases and 73 918 controls, found that smoking significantly increased gastric cancer risk, with an overall odds ratio (OR) of 1.48 (95%CI: 1.28-1.71), and an OR of 1.69 (95%CI: 1.35-2.11) for current smoker status in comparison to never smokers.
Therefore, IM patients with at least 1 of these risk factors (incomplete IM, family history, smoking habit) - information easily available in clinical practice - would appear at a further increased risk of gastric cancer and may probably beneficiate of a more intensive endoscopic surveillance, rather than the 3-year follow-up uniformly suggested for all patients. We recently proposed a patient individualized follow-up with an yearly endoscopic control in those patients with adjunctive risk factors, and a less intensive (2-3 years) follow-up in the remaining IM patients. For instance, a patient with incomplete IM confined to the antral mucosa would not appear to be at lower gastric cancer risk as compared to a patient with extensive, complete IM. Similarly, a smoker patient with complete IM in the antrum or with a family history of gastric cancer likely deserves endoscopic surveillance, contrary to what recommended by the MAPS. However, further studies on this topic are needed.
Finally, the gastric biopsy sampling proposed in the MAPS includes ≥ 2 biopsies in the antrum and ≥ 2 biopsies in the gastric body. However, in the updated Sydney System, 5 biopsies were recommended, including 1 additional specimen on the incisura angularis. The need of this additional biopsy was based on the evidence that IM prevalence is higher in this gastric site as compared to any other part of the stomach. Indeed, IM generally initiates in the incisura angularis, subsequently spreading in both antrum and gastric body. Despite it could be argued that IM only located in the angulus does not increase gastric cancer risk, it is also true that IM would presumably spread to both antrun and body in the majority of patients. Therefore, by simply taking 1 biopsy on the angulus it is possible to early detect IM - that is a generally irreversible, precancerous lesion. Taking 1 further biopsy on the angulus is a simple and rapid procedure, without any additional patient discomfort and cost. Indeed, as suggested in the Operative Link for Gastritis Assessment (OLGA) system, the additional biopsy specimen taken on the angulus should be pushed in the same vial of antral biopsies. The MAPS guideline did not include angulus biopsy among the recommendations. However, the guideline suggested that biopsies should be histologically assessed according to the OLGA/OLGIM system[23,24], for which 5 biopsies (1 on the angulus) are required. Therefore, it would appear reasonable to include incisura angularis in the biopsy sampling.
In conclusion, while waiting for large prospective, randomized, multicenter studies comparing different follow-up strategies, it would appear reasonable to take into account some additional risk factors for gastric cancer in the follow-up strategy for management of gastric precancerous lesions. A simple, patient-tailored surveillance may be probably more appropriate than a single schedule proposed for all patients. Despite MAPS represents a good start, a more patient-orientated road-map(s) could be also considered.