A significant increase in the absolute neutrophil counts (ANC) was observed during treatment with empagliflozin (Figure 2). Increase in ANC was associated with a drop in blood 1,5AG levels (Figure 2A). When blood samples from before (n = 17) and during empagliflozin treatment (n = 20) were compared, a significant increase in the monocyte counts, platelet counts and hemoglobin were also observed during empagliflozin treatment (Figure 2b; p<0.0001 unpaired t test).
To examine changes in neutrophil function, we used a DHR based assay that showed an improvement of the neutrophil SI from 9.46 (before treatment) to a median of 51.9 (during treatment on day 7, 14 and 21) i.e. similar to untreated normal controls (Figure 3A and B).
Neutrophils in patients with G6PC3 deficiency have been extensively studied by Goenka et al.8 who have reported an altered inflammatory activation status of neutrophils in patients with G6PC3 deficiency, possibly due to dependence on glycolysis as their energy source.17 Therefore, we measured surface expression of neutrophil markers18–20 (before and during empagliflozin treatment) to determine neutrophil activation status in our G6PC3-deficient patient. Before treatment, we found higher expression of CD14, CD15, CD18 and CD66B in neutrophils of the G6PC3 deficient patient, similar to the previous report.8 Interestingly, we observed prompt normalization of the mean fluorescence intensity during treatment with empagliflozin – an observation similar to G6PC3 deficient patients that underwent HCT (Figure 3b).8 Thus, there was correction of neutrophil activation during treatment with empagliflozin. Taken together, these results indicate a possible role of empagliflozin in normalizing the neutrophil phenotype by activating glycolysis due to the release of hexokinase inhibition.
The improvement in ANC and neutrophil function was associated with resolution of clinical symptoms. The patient has not developed new infections nor tongue ulcers since starting the treatment with empagliflozin. There has been no episode of abdominal pain or diarrhea during 521 days, since treatment was started. Moreover, our patient reported a significant improvement in the quality of life assessed by PROMIS questionnaire. Cognitive function, emotional distress and clinical symptom scores were below average before treatment. There was significant improvement (T score of > or < 10 i.e. > 1 standard deviation) in all 3 domains during empagliflozin treatment (supplemental table 2). It is noteworthy that the clinical symptoms scores improved to above average for the general population.
There was no evidence of hypoglycemic episodes or any other adverse event during the treatment. Empagliflozin is a frequently used drug in type 2 diabetes with good safety profile.21 The usual starting dose of empagliflozin in an adult with type 2 diabetes is 10 mg daily. To determine the dose of empagliflozin that would be sufficient to treat neutropenia and neutrophil function in G6PC3 deficiency, we started with the daily dose of 5 mg (0.09 mg/kg/day). The dose of empagliflozin was increased to 10 mg/day (0.2 mg/kg/day) on day 42 of treatment, to achieve a normal neutrophil count as shown in Fig. 2A and supplemental table 3. After achieving a normal neutrophil count, on day 52 we reduced the dose to 10 mg every second day (back to 0.09 mg/kg/day) to consistently maintain ANC above 500 x 109/L. Since day 182 our patient has been kept on a maintenance dose of 5 mg every 3rd day (0.03 mg/kg/day). Her neutrophil count has remained consistently above 700 x 109/L on the maintenance therapy for 345 days (since day 182 of treatment) (see Fig. 2A). Patient is asymptomatic without any evidence of infection, tongue ulcers nor gastrointestinal symptoms. It is important to note, that the empagliflozin maintenance dosage (0.03 mg/kg/day) that is used in this patient is more than 10-fold lower in comparison to the previously reported dose (about 0.4 mg/kg/day) used to treat neutropenia in GSD1b patients.11 As yet we do not have an explanation for this, but it is possible that for G6PC3-deficient patients who do not take cornstarch, which is likely an important source of 1,5-anhydroglucitol,22 once they have taken enough empagliflozin to lower the concentration of 1,5-anhydroglucitol in blood (as seen during the initial days of treatment – Figure 2a), a lower maintenance dose can be used to keep 1,5-anhydroglucitol in blood sufficiently low for neutrophils to remain functional and infections to be prevented, as is seen in this particular patient. Clinical trials enrolling more G6PC3-deficient patients treated with empagliflozin are necessary before recommending an empagliflozin dosage for the treatment of neutropenia in G6PC3-deficiency.
Currently, HCT is considered the only curative option for G6PC3 deficiency.9 Typically, HCT is performed in children and adults with G6PC3 deficiency and inflammatory colitis in this GCSF-responsive neutropenia.6–8 Although the neutrophils are GCSF-responsive, they exhibit an activated phenotype and accelerated apoptosis.8 Despite some response to GCSF, patients with G6PC3 deficiency often develop inflammatory colitis. This is not surprising since as shown by Veiga-da-Cunha et al. despite GCSF treatment neutrophils continue to accumulate toxic amounts of 1,5AG6P.10,11 The inflammatory colitis is attributed to abnormal myeloid biology due to G6PC3 deficiency.8 Our patient also had gastrointestinal symptoms, abnormal neutrophil function and counts, and neutrophils with an activated phenotype, before starting empagliflozin treatment. The neutrophil count, function of the neutrophils and their phenotype was normalized, and the gastrointestinal symptoms resolved on empagliflozin. This is an encouraging finding and suggests that reduction in 1,5AG with SLGT2-inhibitor improves the quality of myeloid cells and hence could have a therapeutic role in inflammatory colitis of G6PC3 deficiency. Accordingly, a very recent report, describing role of empagliflozin in the treatment of neutropenia associated with GSD1b, also demonstrated a clear improvement of inflammatory colitis.11
This is the first-in-human report of an SLGT2 inhibitor in G6PC3 deficiency advancing the proof-of-concept demonstrated by G6PC3-deficient mouse models. Complex molecular and genetic techniques are been employed to develop therapeutic options for G6PC3 deficiency and other enzyme deficiencies.23,24 However, the approach of eliminating the toxic metabolite is simple and cost-effective. Similar approaches were tried in the past in a few patients with adenosine deaminase deficiency but the results were not encouraging.25,26 The success of empagliflozin in rescuing the myelopoiesis in G6PC3-deficiency indicates that such simple approaches should be investigated in the laboratory. More trials are now underway to test the definitive role of SLGT2 inhibitor in neutropenia of G6PC3 deficiency and GSD1b. However, our report suggests that patients with G6PC3 deficiency and inflammatory colitis or other end organ damage should be offered the treatment with an SLGT2 inhibitor before proceeding to high-risk procedure of HCT.