Oleic acid chlorohydrin, a new early biomarker for the prediction of acute pancreatitis severity in humans

Acute pancreatitis (AP) is an abrupt inflammatory process of the pancreas that often involves peripancreatic tissues and distant organs [1] and is one of the most common causes of hospital admission due to gastrointestinal disease [2]. The severity of the disease is highly variable, ranging from a mild and self-limited form to a severe disease with local and systemic complications that may lead to multiorgan failure and death [3].

The early prediction of severity becomes a critical issue in the management of AP. When the prediction is accurate, it allows for proper patient stratification with clinical management implications: Patients predicted at risk of developing severe AP will get enhanced hemodynamic support and close monitoring aimed at prompt recognition of local or systemic complications that may require specific treatment. Unfortunately, our ability to identify with high accuracy those severe cases early in the course of the disease is still unsatisfactory. The current prediction of severity is mainly based on multifactorial scores (Ranson, Imrie, APACHE-II, BISAP, etc.) [4, 5]. However, some of them are cumbersome to calculate (APACHE-II, Ranson, Imrie) and, most importantly, these scoring systems have a limited accuracy for mortality or persistent organ failure (> 48 h) [5‐8]. In addition, some reports indicate that these scoring systems are not superior to easier-to-determine biomarkers, such as blood urea nitrogen, C-reactive protein (CRP), hematocrit or creatinine [6]. In this scenario, a blood biomarker with improved accuracy in predicting severe AP (as compared to previously described predictors) would be a step forward toward better early management of patients with severe AP.

In previous studies using an experimental model of taurocholate-induced acute pancreatitis in rats, we observed the generation and release to the bloodstream of specific halogenated lipids [9]. The infiltration and activation of polymorphonuclear neutrophils in the pancreas, but also in surrounding areas of adipose tissue, result in the generation of hypochlorous acid (HOCl) from hydrogen peroxide (H₂O₂) and chloride ions, in a reaction catalyzed by myeloperoxidase (MPO) [10]. The combination of high MPO and pancreatic lipase activity results in the generation and release of fatty acid chlorohydrins (Fig. 1), being the oleic acid chlorohydrin (OAC) the most abundant of them.

It is important to note that the generation of these halogenated fatty acids requires a combination of processes that are triggered during the development of pancreatitis, particularly when the enhanced inflammatory response acquires such characteristics that herald the progression to severe AP. This inflammatory response involves the recruitment of macrophages and neutrophils to the damaged pancreas and to the surrounding adipose tissue. These cells release active MPO in the areas where large amounts of pancreatic lipase catalyze the hydrolysis of triacylglycerols from adipocytes [11]. Therefore, fatty acid chlorohydrins (and in particular OAC) generation is fostered by the combined action of two enzymatic systems (myeloperoxidase and lipase) that are overactive in severe AP. Furthermore, OAC can be released to the bloodstream and its blood levels quantitated.

Experimental studies pointed to an interesting correlation between the levels of circulating OAC and the severity of acute pancreatitis [12]. However, the potential use of plasma OAC concentration as an early biomarker for severe AP in humans has not been evaluated yet. In this study, we assessed the effectiveness and feasibility of OAC in the early prediction of severity of acute pancreatitis in humans.

The relationship between OAC plasma levels and outcome of AP was studied in a prospective cohort of consecutive patients from two medical centers: Hospital General Universitario de Alicante (Alicante, Spain) and Hospital Universitari Vall d’Hebron (Barcelona, Spain). The study protocol was approved by the ethic committee of both centers and followed the Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects. The patients signed an informed consent before entering the study.

Early identification of severe forms of AP is essential for successful management of affected patients. Unfortunately, early reliable and sensitive predictive markers of severity are not at hand for application in clinical practice beyond popular multifactorial scores [22‐24]. Moderately severe-to severe acute pancreatitis is believed to be the result of a multifactorial process that involves tissue necrosis, hydrolytic enzyme activation, release of inflammatory cytokines, generation of reactive oxygen species and synthesis of bioactive lipid mediators. In addition, some of these events take place in extra-pancreatic organs and none of them, by its own, is directly linked to the severity of the disease. Of note, pancreas-derived mediators that proved valuable for severity prediction in experimental animals have failed to predict the future course of pancreatitis when applied to human disease. The concomitant activation of different local and systemic processes explains the need for multifactorial scoring systems developed to predict the severity of AP, including Ranson criteria, APACHE-II or BISAP [5].

Fatty acid chlorohydrins are halogenated lipids generated by the action of hypochlorous acid on unsaturated fatty acids [10]. Although absent in control conditions, they can be released to the bloodstream during AP due to the combined effect of high lipase and MPO activities. Since this combination is characteristic of severe pancreatic/peripancreatic damage and enhanced systemic response, the presence of fatty acid chlorohydrins could only be expected to occur under the simultaneous activation of both hydrolytic and inflammatory processes that, when combined, are associated with the progression from mild to severe AP. Importantly, a number of pro-inflammatory activities have also been reported for fatty acid chlorohydrins, including the induction of P-selectin and the activation of macrophages [9, 25].

The value of lipid metabolites as indicators of increased severity of AP has been progressively recognized. There are also some clues that point to an active role of lipid mediators in the progression of mild to severe AP, in particular on the systemic effects that may lead to organ failure. Obesity or increased intra-abdominal fat is associated with severe AP, possibly through a mechanism related to the lipotoxic action of unsaturated fatty acids released by the effect of lipase on triglycerides of visceral adipocytes. In a series of elegant experiments, Navina et al. [26] demonstrated the pathogenic role of unsaturated fatty acids released during pancreatitis on inflammation, necrosis, multisystem organ failure and mortality. The measurement of chlorohydrins of fatty acids is a further step on this direction since the presence of these molecules indicates that, in addition to the effects of hydrolytic enzymes released by the pancreas, the inflammatory microenvironment required for the halogenation of fatty acids becomes relevant in distant organs. Even when these reactions are only relevant in local scenarios, the plasma levels of OAC are ultimately related to increased severity of AP and to multiorgan failure. The experimental study in rats indicates that release of OAC depends on the action of lipase following a kinetic similar to that observed for oleic acid, whereas its final course seems to be related to the hepatic uptake of free fatty acids [9]. Chlorinated fatty acids have been reported to be catabolized by hepatocytes through ω-oxidation and subsequent β-oxidation and their final products are rapidly excreted in urine [27]. Altogether it suggests that OAC could be a useful prognostic parameter but only in the initial stages of acute pancreatitis.

Our results in human patients confirm the expectations generated in experimental studies and show the value of OAC in predicting the course and outcome of AP. OAC was particularly accurate in predicting persistent organ failure and mortality, the most important outcome variables in AP. Our data reveal that levels of OAC measured on admission allow for discriminating the different degrees of severity to be developed on the progression of the disease in the long run, which should have a direct impact on the clinical management of these patients. According to our data, OAC concentrations higher than 32.4 nM are associated with severe AP, while a cutoff value of 7.5 distinguishes mild from moderate and severe disease.

Like many other studies focusing on research of prognostic biochemical markers, our study also has a number of methodological and technical limitations. The study includes patients from two hospitals, but, as expected, the number of patients in moderate and severe groups was significantly lower than in mild group. Therefore, out of 59 patients with AP only 13 had severe pancreatitis and 16 moderately severe. Although the results obtained allow to discriminate between these three groups with a good statistical level, there is no doubt that more studies with a much higher number of patients need to be done. Anyway, OAC was very accurate to detect severity and was compared with preexisting predictors of severity that were applied to the same sample showing a lower accuracy. On the other hand, the generation of OAC through the incorporation of the HOCl to the 9,10 double bond of oleic acid results in the formation of two isomeric chlorohydrins (the 9-chloro, 10-OTMS 18:0 chlorohydrin and the 9-OTMS, 10-chloro 18:0 chlorohydrin). These compounds are not separated by the GC column and, although this limitation may have little impact on the overall results of our study, we have to acknowledge that our data indicate the sum of the two isomers. In addition, it must be taken into account that, although the measure of OAC by gas chromatography/mass spectrometry is too complex and slow to be applied to clinical practice, ELISA tests similar to those for prostaglandins and other lipid metabolites can be easily developed allowing for much faster measurements of OAC. Availability of faster and easier assays methods will make easier to confirm our results in larger samples of patients. In this line, although these encouraging results have been obtained from two independent cohorts of patients from two different and geographically remote hospital centers, they should be validated in larger cohort studies performed by other groups. In case of similar results, OAC would be, by far, the most accurate early predictor of severe acute pancreatitis ever described.

In summary, this study demonstrates that OAC is generated during AP, it can be measured in plasma, and its levels correlate with pancreatitis severity. Our findings indicate that oleic acid chlorohydrin is an accurate early prognostic biomarker to be used in patients with acute pancreatitis.