Poster Number-27 - Potent anti-inflammatory effects of an H2S-releasing naproxen (ATB-346) in a human model of inflammation

Potent anti-inflammatory effects of an H₂S-releasing naproxen (ATB-346) in a human model of inflammation

Purpose:

Inflammation is a protective response against infection or injury. It is commonly treated with non-steroidal anti-inflammatory drugs (NSAIDs), including naproxen. The long-term use of such drugs is limited by significant adverse effect. Of note, NSAIDs increase cardiovascular risk, as well as causing unacceptably high levels of gastrointestinal (GI) disturbance, with mucosal irritation and even ulcer formation, often leading to cessation of drug use. Such GI effects may, to an extent, be ameliorated by the co-administration of gastroprotective drugs, but given the growing problem of polypharmacy and the propensity for untoward drug interactions, clearly single agents are preferable. Thus, there is a need to develop more efficient and effective therapeutics. Further, given the ongoing opioid crisis, non-addictive, efficacious alternatives are becoming increasingly relevant.

In this regard, attention has turned to an emerging class of compounds with significant anti-inflammatory effects, the hydrogen sulfide (H₂S)-releasing NSAIDs (H₂S-NSAIDs). These consist of conventional NSAIDs to which an H₂S-reasing moiety is covalently attached. One such H₂S-NSAID, ATB-346, has shown markedly reduced GI adverse effect profile (in terms of ulceration and bleeding) in animal studies. Furthermore, it was shown that after oral administration of ATB-346, plasma levels of naproxen derived from the compound were much lower than equimolar doses of naproxen but exhibited cyclooxygenase (COX) inhibition and analgesic effects.  ATB-346 is in the order of six-times more potent than naproxen in humans and is as effective at suppressing COX activity over a 24-hour period with a single dose (versus twice-daily naproxen). In a recent Phase 2b trial in healthy volunteers, the incidence of upper GI ulceration in subjects taking standard dose twice daily naproxen was over 42%; significantly greater than the rate of ulceration of only 2.5% (p< 0.001) in subjects taking an equi-effective dose of ATB-346 daily. 

Using a novel model of ultraviolet (UV)-killed Escherichia coli (UV-KEc)-triggered resolving dermal acute inflammation, we set out to investigate the anti-inflammatory role of ATB-346 in humans. Comparing this to its native counterpart, naproxen, and untreated controls we aimed to evaluate whether next generation H₂S-NSAIDs, whilst demonstrating lower GI adverse effect, are still able to retain their anti-inflammatory effect in humans.

Methods:

Twenty-one healthy, male volunteers aged 18-50 were recruited. Volunteers were randomly allocated to one of three treatment arms in this single-blind study. Seven volunteers were assigned to the naproxen arm and took 500 mg twice daily for three days prior to UV-KEc injection. Seven volunteers were assigned to the ATB-346 arm and took 250 mg once daily (the equi-effective dose to naproxen) for three days prior to injection.  Seven volunteers took no drug during the study and served as the untreated control group.

Into the forearm of each volunteer, 1.5 x 10⁷/100 µL UV-KEc in 100 µL of 0.9% sodium chloride were injected intradermally. Inflammation was allowed to ensue for the duration of the study with clinical measurements, laser Doppler imaging and peripheral venous blood taken at pre-defined time-points representing baseline (0 hr), onset of inflammation (4h) and resolution phases (24h and 48h). At each time point, peripheral blood was taken for complete blood count (CBC) and C-reactive protein (CRP). Laser Doppler imaging was used to measure vascular hyper-reactivity. The scanner emits a laser which is scattered by erythrocytes with the resulting Doppler shift dependent on the velocity and concentration of cells at the site, therefore representing blood flow and thus leucocyte trafficking, a hitherto unappreciated marker of resolution of inflammation. As a further parameter of inflammation (and thus resolution), a visual analogue score (0-10) was used to quantify the pain experienced at the site of inflammation (0 = no pain; 10 = worst pain imaginable). This was repeated for elicited tenderness with the application of a 100 g weight. Temperature was measured using an electronic thermometer both centrally (forehead) and at the site of the injection.

Inflammatory exudate was obtained from the site of the UV-KEc injection by formation of a suction blister at two time-points; 4h and 48h. Blister cells were incubated with an antibody cocktail along with appropriate fluorescence minus one (FMO) controls and acquired immediately using a flow cytometer. Finally, the blister supernatant was diluted and prostaglandin E₂ (PGE₂) quantification, by enzyme-linked immunosorbent assay (ELISA) and for interleukin-10 (IL-10) and TNF-α quantification, a customised two-plex array.

Results:

There were no significant differences in age (mean 25 years) or ethnicity between treatment groups. No adverse effects were reported in any group. Approximately 20% of total volunteers complained of axillary heaviness between 4h and 24h, associated with linear, erythematous tracking from injection site towards the axillae; likely representing pain-free, lymphatic drainage. The mean blister volume at 4h was 129.73 µL and 135.43 µL at 48h. No significant differences were seen in peripheral blood.

Cell profiles were characterised using flow cytometry. Neutrophils were defined as Lineage^- (Lin^-: CD3, CD19, CD20 and CD56), HLA-DR^-, CD16⁺, Siglec-8^-. Mononuclear phagocytes (MP) were identified as Lin^- and HLA-DR⁺. Classification into MP subtypes was based on expression of CD14 and CD16: classical monocytes, CM (CD14⁺, CD16^-), intermediate monocytes, IM (CD14⁺, CD16⁺), non-classical monocytes, NCM (CD14^-, CD16⁺) and dendritic cells (CD14^-, CD16^-). Lin⁺ cells were further divided into B cells (HLA-DR⁺) and T/NK cells (HLA-DR^-).

ATB-346 significantly reduced neutrophil numbers at the peak of onset of inflammation (4h). In the naproxen group, there was a significant reduction in neutrophil infiltration at 4h compared to untreated, albeit to a lesser extent than ATB-346. At 48h there were no significant differences in neutrophil numbers. HLA-DR^- T/NK cells were significantly lower in those treated with ATB-346 compared to untreated at 48h. CM and IM numbers were increased at 48h. ATB-346 and naproxen caused a trend to reduced numbers of CM and IM at 48h. At 48h, both CM and IM exhibited significant reduction in expression of CD14 based on geometric mean fluorescence intensity in both naproxen and ATB-346 treated groups, compared to untreated. No significant changes were seen in other cell populations.

Volunteers treated with ATB-346 reported significantly lower pain and tenderness scores at the time-point of maximal neutrophil infiltration (4h) compared to untreated; effects also observed with naproxen. Temperature at the site of inflammation followed the same course in ATB-346 and naproxen groups, peaking at 24h and at 4h in the untreated. Vascular hyper-reactivity peaked at 24h in all groups declining back to similar levels seen at 4h by 48h. At all time-points, there was a trend towards increased hyper-reactivity in the ATB-346 group compared to naproxen and untreated.

While there was a trend towards an increase in TNF-α and reduction in IL-10 in both treatment groups, these were not significant. There was a significantly reduced concentration of PGE₂ in both ATB-346- and naproxen-treated volunteers compared to untreated.

Conclusion:

We have shown that ATB-346, is potently anti-inflammatory, as evidenced by a significant reduction in neutrophil numbers at 4h compared to untreated controls. We observed a significant reduction in the expression of CD14 on CM and IM in both drug treatment groups compared to untreated controls at 48h, which could represent a switch to a more anti-inflammatory phenotype given its role in LPS-induced monocyte activation and role in fighting infection.

Both ATB-346 and naproxen significantly reduced pain and tenderness at 4h compared to untreated, the time-point of maximal neutrophil infiltration, suggesting a reduction in numbers of inflammatory cells mediators may contribute to the analgesic effect of both drugs.

As the opioid crisis continues in the developed world, it is imperative that we find novel, efficacious therapeutics that are agreeable to patients in terms of adverse effects and risk. Whilst more work is needed to explore the mechanisms underpinning the reduction of neutrophils seen in those treated with ATB-346, we propose it represents the next generation of H₂S-NSAIDs as a viable alternative to conventional NSAIDs, showing reduced GI adverse effects, with actions through multiple potential facets of the immune system and a novel mode of action mediated through the H₂S moiety.

References: