A phase 4, monocenter, randomized, double-blind, placebo-controlled, 4-armed cross-over mechanistic intervention study to assess the renal hemodynamic effect of mono- and combination therapy with empagliflozin (SGLT-2 inhibitor) and losartan (RAS inhibitor) in metformin and/or SU-treated patients with type 2 diabetes mellitus

Diabetic kidney disease (DKD), characterized by reduced whole-kidney glomerular
filtration rate (GFR) and/or urinary protein leakage, is a feared complication
of type 2 diabetes (T2D). With severe consequences such as endstage kidney
disease (ESKD) and renal death, and strongly linked to cardiovascular (CV)
morbidity and mortality, optimal treatment of DKD is vital. Still, even with
multifactorial treatment of renal risk factors, including hyperglycemia,
hypertension, obesity, dyslipidemia and albuminuria, residual risk remains high
worldwide. Since the introduction of blockers of the
renin-angiotensin-aldosteron system (RAS), no other renoprotective drug for T2D
has been successfully developed, highlighting the need for novel strategies or
new therapeutic drugs to improve renal outcome in T2DM.
In this regard, the introduction of the sodium glucose cotransporter (SGLT)2
inhibitors has been met with great enthusiasm. Designed to inhibit glucose
reabsorption in the proximal tubule they induce glycosuria which indeed reduces
hyperglycemia. More importantly, these drugs have shown remarkable benefits on
CV disease and renal outcome in large CV safety trials in T2D patients with
high risk of or established atherosclerotic cardiovascular disease (CVD) as
well as in patients with DKD. The first of these trials, the EMPAgliflozin
cardiovascular outcome event trial in type 2 diabetes mellitus patients*
Removing Excess Glucose (EMPA-REG OUTCOME), was reported in 2015 and
demonstrated, next to risk reductions in CV outcomes, impressive reductions in
the prespecified secondary renal outcome (1). In two subsequently reported CV
safety trials conducted with canagliflozin (CANVAS-Program) and dapagliflozin
(DECLARE-TIMI 58), these promising results indicating renal benefit were
further strengthened (2,3). Recently, the results of a dedicated
placebo-controlled trial with canagliflozin (CREDENCE) in DKD patients were
reported (4). The study was terminated early due to overwhelming beneficial
effects. Yet, at this point in time, the renoprotective mechanisms involved
with SGLT2 inhibition still remain speculative, though a consistent finding is
that SGLT2 inhibitors reduce estimated GFR after first dosing, which is
reversible after treatment cessation. This *dip* indicates a renal hemodynamic
phenomenon reminiscent of the RAS blockers and is thought to reflect a
reduction in intraglomerular pressure (5). From studies in rodent models of
type 1 diabetes and humans with type 1 diabetes it is hypothesized that SGLT2
inhibition leads to urinary sodium excretion by inhibiting in the proximal
tubule, which influences renal hemodynamics through a mechanism known as
tubuloglomerular feedback. In short, reduced sodium reabsorption at the level
of the proximal tubule leads to increased sodium chloride delivery at the
downstream located macula densa, which in turn increases afferent arteriolar
resistance and reduces glomerular (hyper)filtration and hydrostatic pressure
(5). In a recent trial at our department RED (NCT02682563) we investigated
whether this is also true in T2D. To our surprise, this study showed that the
renohemodynamic actions of SGLT2 inhibition in T2D are not due to afferent
vasoconstriction but rather efferent vasodilation. This is also the proposed
working mechanism of inhibitors of the RAS system in T2D, although dedicated
studies in humans are scarcely done. Indeed, people with T2D that do not
respond to RAS blockers in terms of albuminuria reduction, also do not respond
to SGLT2 inhibitor treatment (6).
In our opinion, we are left with several questions regarding the combination of
SGLT2 and RAS inhibitors. Especially with the recent results of CREDENCE, it is
very likely that the combination of these agents will become standard of care
in patients with T2D and DKD. Both agents dilate the postglomerular arteriole,
which might lead to relevant interactions or even synergistic effects. Since
the majority of the population in the cardiovascular outcome trials used RAS
inhibition, we know the renoprotective effect of SGLT2 inhibition is present
with concurrent RAS inhibition. However, to what extend these agents interact
and which of the various complex pathways involved in blood pressure and plasma
volume control are affected by mono or combination therapy with these agents is
unknown. It is important to emphasize that in the large trials, RAS blockade
was not randomized and that the participants not on RAS blockade were small in
numbers, making additional analyses on this topic difficult.
In conclusion: Despite multifactorial treatment approaches, residual risk for
the development and progression of DKD remains high, and novel therapies or
strategies to halt renal burden in T2DM are urgently needed. SGLT2 inhibitors
and RAS inhibitors both induce glucose-independent renoprotective effects and
improve renal outcome, seemingly via an at least partly equal mechanism, the
dilation of the efferent glomerular arteriole resulting in an eGFR dip. The use
of combination therapy with these agents could lead to an additive or even
synergistic renoprotective effect in T2DM. As such, combined use of an SGLT2
inhibitor and RAS inhibitor may enhance individual benefits (e.g. reduction of
glomerular pressure, activation of tubuloglomerular feedback, proximal and
distal natriuresis, plasma volume contraction and reduction of blood pressure).

Investigate the effects of 7 (+-1) days of of mono- and combination therapy
with the SGLT2 inhibitor empagliflozin (10 mg QD) and RAS inhibitor losartan
(50 mg QD) on renal hemodynamics, (glomerular filtration rate (GFR) / effective
renal plasma flow (ERPF)) in metformin and/or SU-treated T2DM patients.

A single-center, prospective, placebo-controlled, double-blind, randomized,
4-arm cross-over mechanistic intervention study in 24 metformin and/or
SU-treated patients with T2DM.

24 T2DM patients will be randomized 1:1:1:1 to receive either empagliflozin and
losartan, or empagliflozin and placebo, or losartan and placebo, or placebo and
placebo.

Participants will be randomized for a four times 7 (+-1) days treatment with
two different active study agents. Consequently, independent of treatment
allocation, beneficial effects can be expected, as SGLT-2 inhibitors improve
glycemic control and RAS inhibitors lower blood pressure. All study medications
have been approved for blood-glucose lowering or hypertension treatment in T2DM
and hypertensive patients and, based on currently available data, are
considered to be safe. Furthermore, SGLT-2 inhibitors in general may also
decrease blood pressure and body weight. It is important to stress that the
patients have suboptimal glycemic control and an indication to receive
additional glucose lowering medication.
The most common adverse effects for empagliflozin are genital mycotic- and
urinary tract infections, pruritus, polyuria, frequent voiding and nycturia.
When used in combination with a SU derivative or insulin (which is not the case
in our study) hypoglycemia can occur. In addition, a slight
empagliflozin-induced increase in LDL-cholesterol and increased hematocrit have
been reported. Long-term (adverse) effects of SGLT-2 inhibitors are currently
under investigation in large-scaled outcome trials, however the
SGLT2-inhibitior empagliflozin has shown to reduce mortality and heart and
renal failure in the EMPA-REG OUTCOME Trial (Bernard Zinman et al., 2015) (B.
Zinman et al., 2015) and the SGLT2-inhibitior canagliflozin showed similar
results in the more recent CANVAS trial (Neal et al., 2017).

With regard to the used testagents, the Infusion of Iohexol can lead to a warm,
sometimes painful sensation. Most common but rare adverse effects are headache,
stiffness, nerve pain, nausea, vomiting, fever, hives, stomach pain,
hallucinations and neurological changes. In patients with an allergy for iodide
it can elicit hypersensitive reactions, therefore we specifically check a for
this allergy at screening.

As in all drug intervention trials, in this study, we will closely monitor
patients for adverse drug and study events during the follow-up visit and by
telephone consultation according to GCP (see Appendix A). Participants can
contact the research staff 24 hours a day.

Possible benefits for participants
Participants are extensively tested and monitored, and receive high-quality
care with many contact moments without any costs. In addition, this study is
ideal to futher detail the effects of mono- and combination therapy with the
SGLT-2 inhibor empagliflozin and RAS inhibitor losartan on renal hemodynamics.
The patient has the opportunity to gain experience with these agents at no cost
and might benefit from it later (when intensification of therapy is necessary).
In addition, former studies have shown that the use of SGLT2-inhibitors has a
beneficial effect on stabilization of GFR and the lowering of blood pressure
and weight.

Possible inconvenience for participants
Over the last 10 years, we have gained ample experience with similarly
demanding mechanistic drug intervention studies in T2DM patients on renal
hemodynamics (SAFEGUARD 2012.391, RENALIS 2013.459, ELIXIRS 2014.275, RED
2015.421, RACELINES 2017.336). Based on the positive feedback from our
participants, the low drop-out rate (max 5%) and the large proportion of
participants that returns to participate in yet another (similarly demanding)
study, we are confident that the burden on participants is perceived as not
being too high. Indeed, we have built in different ways to alleviate the burden
for participants, including clear, repeated communication, frequent contacting,
intensified (diabetes) care, 24-hour availability of research staff, study and
travel reimbursement, enabling participants to receive the newest study
medication (that for most of them would not be reimbursed in daily practice)
and offering follow-up care in our out-patient clinic. Finally, it should be
noted that several tests are similar to currently or previously performed in
patient care for diagnostic purposes (e.g. iohexol/PAH-clearance).
We are aware of the fact that in the current study participants will undergo
multiple tests that demand a considerable time investment from their end. The
total duration of visits is 2 hours (V1) and 16 hours (visit 2, 3, 4, 5). These
renal / cardiovascular test-days may be perceived as demanding that amongst
others involves frequent blood and urine collection, infusions, blood pressure
and heart rate. As mentioned above, all possible measures will be taken to
minimize the discomfort for the participants during the tests (e.g. comfortable
beds are available which allow a semi-recumbent position).

Safety issues
The study examinations/tests are considered to be safe. No invasive procedures
(besides intravenous peripheral catheters) are involved. During the study
tests, two *diagnostic agents* (i.e. iohexol and PAH) need to be administered;
both agents are inert and have no side effects.