Hydroxyl Free Radical Associated Effects of SGLT-2 Inhibitors in Type 2 Diabetes

Distal symmetric polyneuropathy (DSPN) is one of the most common microvascular complications of type 2 diabetes mellitus (T2D), affecting approximately 50% of patients over the course of their disease. This complex disorder involves both metabolic and vascular insults leading to nerve fiber damage with hyperglycemia-driven oxidative stress, reduced nerve blood flow, and chronic subclinical inflammation contributing to its pathogenesis.

The presence of DSPN impairs quality of life and is associated with increased mortality as well.

Despite achieving optimal glycemic control, diabetic neuropathy frequently progresses due to multifactorial influences, including chronic oxidative stress and microvascular injury. Traditional therapies have focussed on symptomatic relief, such as pain management with anticonvulsants or antidepressants, and on attempts to slow disease progression by targeting hyperglycemia. Consequently, new interventions addressing additional pathogenic factors, including oxidative stress, could offer improved clinical outcomes in DSPN. In the aetiology of diabetic neuropathy, oxidative stress plays a significant role.

Under physiological conditions, phenylalanine is hydroxylated enzymatically at the para position, producing para-tyrosine. However, under conditions of increased oxidative stress, for instance during hydroxyl radical (•OH) attack, phenylalanine can be mis-hydroxylated at the ortho and meta positions, giving rise to ortho-tyrosine (o-Tyr) and meta-tyrosine (m-Tyr). These atypical tyrosine isomers are typically negligible in healthy states; thus, their elevated levels or increased abnormal ratios serve as sensitive biomarkers of hydroxyl radical-mediated oxidative damage, which may be particularly relevant in chronic diseases associated with persistent oxidative stress. Given these oxidative stress pathways, alpha-lipoic acid (ALA) has been explored as an antioxidant therapy that primarily targets free radical production.

ALA has been utilized as a parallel or adjunctive approach to existing glycemic management strategies, with short-term trials showing significant improvements in nerve conduction velocity and neuropathic symptoms. At the same time, ALA therapy as a systemic antioxidant treatment, at the same time may have a beneficial effect on the cardiovascular hard endpoints, as we proved previously.

While ALA therapy underscores the importance of tackling oxidative mechanisms, sodium-glucose cotransporter-2 (SGLT-2) inhibitors are currently under investigation for potentially broader protective effects.

SGLT-2 inhibitors exert their primary effect in the proximal tubule of the kidney by blocking the reabsorption of glucose and sodium, resulting in insulin-independent glycemic control through enhanced urinary glucose excretion. This process induces a mild osmotic diuresis and natriuresis, thereby reducing blood volume and leading to clinically relevant decreases in blood pressure, typically in the range of 2-5 mmHg. Additionally, SGLT-2 inhibitors have been shown to mitigate oxidative stress and inflammation, i.e., mechanisms that may underline the cardiovascular and renal protective effects observed.

Animal experiments have confirmed improved nerve conduction deficits and prevention of intraepidermal nerve fibre loss with SGLT-2 inhibition in diabetic rat models, presumably through reduced glycemic variability, improved microvascular function, and modulation of hydroxyl radical pathways.

Large cardiovascular trials (e.g., EMPA-REG, CANVAS, DECLARE) have primarily focussed on cardiovascular endpoints rather than neuropathy-specific metrics. In a post hoc analysis of EMPA-REG OUTCOME, changes in hematocrit alone were estimated to account for approximately 52% of empagliflozin's cardiovascular benefit.

Although other major SGLT-2 inhibitor trials (DECLARE-TIMI 58, CANVAS, EMPA-KIDNEY) did not provide a breakdown of similar percentage, they each reported hematocrit elevations that may reflect a possible erythropoietin-like effect - an effect further highlighted in DAPA-CKD, where increased hematocrit was thought to be a key factor in the nephroprotective benefits of the drug.

On the other hand, a small, mechanistic study proved a temporary elevation of the erythropoietin serum level using canagliflozin, which was present in the first 4 weeks of SGLT-2 therapy only, while the increase of hematocrit was prolonged for the whole, 12 weeks long treatment period. Although there are other studies either suggesting longer duration of erythropoietin elevation even for 12 weeks by empagliflozin, or no increase of it at 24 weeks by dapagliflozin, both trials verified a retained increase in RBC-parameters for the whole treatment period by the SGLT-2 inhibitors.

However, to this date, few studies have systematically evaluated the short-term neuroprotective impact of SGLT-2 inhibitors alongside changes in hydroxyl radical markers and RBC parameters in a single cohort.

Notably, a comprehensive systematic review and meta-analysis by Kandeel, involving six human studies and a total of 5,312 diabetic patients, demonstrated that SGLT-2 inhibitors moderately improved diabetic peripheral neuropathy symptoms, including improved sensory and motor nerve conduction velocities and potential autonomic stabilization.

In support of these findings, Ishibashi et al. reported that three years of SGLT-2 inhibitor therapy in patients with modestly controlled type 2 diabetes significantly improved neuropathy outcomes and decelerated nephropathy progression. In addition, recent perspectives emphasize that early intervention and addressing multiple metabolic factors may stabilize or even reverse nerve damage in DSPN, and short-term clinical evidence shows that empagliflozin can improve electrophysiological parameters, reduce oxidative stress markers, and moderate neuropathic symptoms in type 2 diabetic patients.

These findings point to mechanisms extending beyond glycemic control, encompassing RBC formation, changes in hydroxyl radical metabolites (such as ortho- and meta-tyrosine), and renal hemodynamics.

Given this encouraging data, we tested a 4-week intervention to capture immediate RBC and hydroxyl radical changes as well as early neuropathic shifts.

Given the limited efficacy of existing treatments, these potential protective effects are particularly salient in the context of diabetic neuropathy's complexity and, by directly or indirectly reducing oxidative stress and influencing pathways that underlie nerve damage, SGLT-2 inhibitors could herald a paradigm shift in the management of this challenging complication.

Against this background, our study focuses on evaluating the short-term effects of SGLT-2 inhibitor (dapagliflozin or empagliflozin) therapy on diabetic neuropathy-related measures, hydroxyl radical markers, abnormal albuminuria, and RBC-parameters in patients with type 2 diabetes mellitus. Specifically, we aim to determine whether a four-week SGLT-2 inhibitor therapy can produce measurable changes in neuropathic function among T2D patients with early or mild abnormalities and to review how shifts in hydroxyl radical markers relate to these clinical and laboratory findings.

Materials and methods

Study population, inclusion/exclusion criteria A total of 40 adult T2D patients were enrolled to the study. Eligible participants included those for whom, in addition to administering any existing antidiabetic regimen, the treating physician decided to initiate SGLT-2 inhibitor therapy (empagliflozin or dapagliflozin) based on standard clinical indications.

The study included T2D patients aged 18 to 75 years with HbA1C<10% and eGFR>30ml/min/1.73 m2. Out of the 40 patients, none reported painful neuropathy (DN4 questionnaire), 9 exhibited neuropathic sensory symptoms (NTSS questionnaire), and 9 showed loss of protective sensation (monofilament test). According to the Neuropathy Disability Score (NDS), 14 patients had mild neuropathy, 16 had moderate neuropathy, and 4 had severe neuropathy, indicating that most patients had preexisting asymptomatic neuropathy. Patients were excluded in case they had neuropathies of non-diabetic etiology (i.e., hereditary, drug- or toxin-induced, untreated hypothyroidism, folate deficiency (below 3.9 µg/L), vitamin B12 deficiency (below 200 ng/L), advanced hepatic- renal disease not meeting the inclusion criteria, autoimmune disease, malignancy, or if they were engaged in regular excessive alcohol consumption (more than two units in men and more than one unit per day in women - one unit being defined as 15 g of pure alcohol). Patients with severe peripheral arterial disease, a history of foot ulceration, or recurrent genital infections were also excluded.

Study design and objectives This prospective, open-label, single-arm clinical study aimed to investigate the short-term effects of sodium-glucose cotransporter-2 (SGLT-2) inhibitors on distal symmetric diabetic polyneuropathy, hydroxyl radical markers, and related hematological and renal parameters in patients with type 2 diabetes mellitus (T2D). The rationale was based on the putative mechanism whereby, beyond their glucose-lowering effect, SGLT-2 inhibitors may influence hydroxyl radical marker formation and thereby improve microvascular complications. This includes potential benefits on red blood cell (RBC) parameters, abnormal albuminuria, and early-stage diabetic neuropathy. The primary hypothesis was that short-term SGLT-2 inhibitor therapy would mitigate hydroxyl free radical pathways specifically tyrosine metabolite ratios accompanied by improved neuropathic measurements and RBC-related indices.

Intervention and study protocol All participants underwent a baseline (week 0) assessment followed by a follow-up evaluation after 4 weeks of SGLT-2 inhibitor therapy. The dose and choice of SGLT-2 inhibitor were determined by the clinical team based on standard treatment indications. No additional interventions were mandated by the study.

Data collection and questionnaires Participants completed lifestyle and medication questionnaires at baseline, documenting diet, exercise, smoking and alcohol habits, concomitant medications, and comorbidities. The Neuropathic Total Symptom Score (NTSS) and DN4 questionnaires were administered to evaluate subjective neuropathic complaints.

Neuropathic assessments Neuropathy evaluation was performed at the Neuropathy Center of the 2nd Department of Medicine and Nephrology-Diabetes Center, University of Pécs Medical School, using the CA-12 software-based system (MSB-MET Ltd., Balatonfüred, Hungary). Quantitative sensory testing was conducted to measure current perception thresholds (CPT) at three frequencies (2000 Hz, 250 Hz, and 5 Hz) on both median and peroneal nerve, providing insight into large myelinated (2000Hz Aβ), medium-size myelinated (250Hz Aδ) and unmyelinated (5Hz C) small fibers function. CPT values below the normal range indicate hyperesthesia, and values above the normal range indicate hypoesthesia.

Additional clinical examinations were conducted for distal symmetric polyneuropathy (DSPN):

1. Deep tendon reflex testing, and 128 Hz calibrated tuning fork (vibration) for large fibre dysfunction, and
2. Pinprick (pain), and temperature perception for small fiber dysfunction. The 10 g monofilament (protective sensation) evaluation for the risk of developing foot ulcers DSPN is one of the most common complications in patients with type 2 diabetes (T2D), even with well-controlled glycemia. DSPN can affect either small or large nerve fibers or a combination of both in a length-dependent manner, in a "stocking-glove" distribution.

Laboratory measurements Blood and urine samples were collected at baseline and at week 4. No additional venipuncture beyond routine sampling was required; approximately 10 ml of blood was obtained for research purposes.

Routine parameters The HbA1c was measured at week 0. Serum fructosamine, fasting glucose, and serum insulin were measured at both week 0 and week 4. Serum creatinine and estimated glomerular filtration rate (eGFR) were assessed at week 0 and week 4. Serum bilirubin, AST, ALT, GGT, ALP, LDH, INR, serum total protein, albumin, total cholesterol, HDL, LDL, uric acid serum iron, transferrin, transferrin saturation, and ferritin were measured at week 0, whereas triglycerides, serum sodium, potassium, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and urinary albumin and creatinine were measured at both week 0 and week 4.

Hydroxyl free radical markers To explore the hydroxyl free radical pathway and its link to neuropathy, RBC-formation, and renal parameters, markers were measured at the baseline and at week 4. Serum and urinary phenylalanine (Phe), para-tyrosine (p-Tyr), meta-tyrosine (m-Tyr), and ortho-tyrosine (o-Tyr), were quantified by high-performance liquid chromatography (HPLC). In addition to the direct concentrations of these metabolites, the following ratios were calculated to characterize changes in the oxidative milieu under SGLT-2 inhibitor therapy: p-Tyr/Phe, m-Tyr/Phe, o-Tyr/Phe, (m-Tyr+o-Tyr)/Phe, m-Tyr/p-Tyr, o-Tyr/p-Tyr, (m-Tyr+o-Tyr)/p-Tyr, m-Tyr/krea, p-Tyr/krea, o-Tyr/krea, as well as any corresponding serum and urinary values. This comprehensive panel of hydroxylation products and derived ratios aimed to provide a detailed assessment of hydroxyl free radical activity and its potential influence on diabetic complications.

Additional calculations The Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) was calculated to assess insulin sensitivity. Changes in the urinary albumin/creatinine ratio (ACR) and correlations with current perception threshold (CPT) parameters were also evaluated throughout the study.

The Neuropathy Disability Score (NDS) was calculated using a standardized clinical examination approach. This assessment evaluates ankle reflexes, vibration sensation, pin-prick sensation, and temperature sensation at the big toe. The maximum total score is 10, categorized as follows:

3-5 points: mild neuropathy 6-8 points: moderate neuropathy 9-10 points: severe neuropathy

High performance liquid chromatography (HPLC) analyses Venous blood was taken from patients into native tubes during routine blood sampling. Serum was extracted by centrifugation and was stored at -80°C pending further examination. Thereafter, 125 µl of trichloro-acetic acid (TCA) was added to 500 µl of serum, then samples were incubated on ice for 30 minutes. Afterwards, precipitate was separated by centrifugation and supernatant was filtered using a 0.2 μm syringe filter (Millipore Inc.). Supernatant was stored at -80 °C until undergoing further process or was immediately analyzed by HPLC. HPLC analysis was carried out using a Shimadzu Class LC-10 ADVP HPLC device (Shimadzu Manufacturing Inc., USA). A 20 μl of filtrate was injected onto a Licrospher-C18 silica column (Merck). Different amino acids and isomers: phenylalanine (Phe), para-tyrosine (p-Tyr), meta-tyrosine (m-Tyr), ortho-tyrosine (o-Tyr) were detected using their autofluorescence. p-, m-, o-Tyr were measured at 275 nm excitation and 305 nm emission, while Phe was evaluated at 258 nm excitation and 288 nm emission using a Shimadzu RF-10AXL fluorescent detector. An isochratic flow of the eluent containing 1% acetic acid, 1% sodium acetate and 98% distilled water was performed. For determination of the concentrations of different amino acids and isomers, external standards were used.

Statistical Analysis Data were analyzed by a biostatistician using appropriate methods for paired comparisons as well as correlation and regression analyses to identify potential predictors of change in neuropathy, hydroxyl radical-related and metabolic parameters. Subgroup analyses reviewed patients with abnormal baseline CPT values to discern whether SGLT-2 inhibitor therapy might confer greater neurological benefits in those with pre-existing neuropathic alterations. Distribution of parameters was tested using Kolmogorov-Smirnov's normality test. Changes of parameters between baseline and week 4 were compared using paired sample t test.

Correlational analyses were performed by Pearson's (normal distribution) or Spearman's rho (non-normal distribution) tests. Linear regression models with stepwise method were established for evaluating predictors of neuropathy, RBC parameters and albuminuria. The Statistical Package for Social Sciencies (SPSS) version 28 (IBM, Armonk, New York, United States) was used for statistical analyses. A 'p' value below 0.05 was considered as statistically significant.

In summary, the methods were designed to investigate whether short-term SGLT-2 inhibitor therapy, by reducing hydroxyl free radical generation and improving RBC indices and abnormal albuminuria, could also yield measurable improvements in diabetic neuropathic function and related hydroxyl free radicals.