November 11-15, 2017; Anaheim, CA; Day #3 Highlights – Draft

Executive Highlights

We’re back with more coverage from AHA 2017, going on now in sunny Anaheim, California. Day #3 was absolutely packed with diabetes content. We saw new analyses of CANVAS and EXSCEL, investigating differential effects of diabetes drugs in primary vs. secondary CV prevention. We visited more diabetes booths than ever before for an AHA meeting in the exhibit hall. We heard a rousing call-to-action for cardiologists to get more actively involved in diabetes care. We learned the latest blood pressure guidelines (published Monday), <130/80 mmHg for people with diabetes. Notably, two of the nine highlights on this page focus on patients with peripheral arterial disease (PAD): a post-hoc analysis of EMPA-REG OUTCOME and a post-hoc of FOURIER. We’re intrigued by this emphasis on PAD as a diabetes comorbidity, with growing recognition that it confers added CV risk on top of what’s already heightened CV risk for people with diabetes. Dive into all our day #3 highlights below!

Click here for a look back at days #1-2 of AHA 2017, and here for our conference preview (there are still two days of this fascinating meeting to go.)

Top Nine Highlights

1. Stanford’s Dr. Kenneth Mahaffey discussed a subgroup analysis of CANVAS suggesting consistent CV and renal benefits to Invokana (canagliflozin) therapy across primary and secondary prevention cohorts. The data was simultaneously published online in Circulation. In her discussant, Oxford’s Dr. Angelyn Bethel was skeptical that Invokana’s benefits to three-point MACE apply in a primary prevention population (she posed important questions related to mechanism to build her argument), but she expressed more confidence that the drug’s heart failure and renal benefits do translate. We also include below insights from our interview with Dr. Mahaffey.

2. Dr. Subodh Verma presented a new post-hoc analysis of EMPA-REG OUTCOME (for Lilly/BI’s SGLT-2 inhibitor Jardiance) that focused on patients with baseline peripheral arterial disease (PAD). Safety and efficacy findings in this subgroup were comparable to the overall CVOT results, and Dr. Verma underscored that empagliflozin was not associated with any increase in lower limb amputations. Notably, in his discussant, Dr. Renato Lopez cautioned against over-comparison between CANVAS and EMPA-REG OUTCOME – he emphasized that amputation was not formally adjudicated in either trial.

3. Presenting a new post-hoc analysis from EXSCEL, Duke’s Dr. Robert Mentz showed consistent effects to AZ’s GLP-1 agonist Bydureon (exenatide) on all-cause mortality and three-point MACE regardless of baseline risk quintile. He highlighted non-significant p-values for interaction – 0.2 for all-cause death and 0.79 for the primary outcome of three-point MACE (non-fatal MI, non-fatal stroke, or CV death).

4. The release of new blood pressure guidelines from ACC/AHA generated quite a lot of buzz around the conference center on Monday: With the threshold for hypertension moved down from 140 mmHg systolic to 130 mmHg systolic, nearly half the US adult population could now be diagnosed with high blood pressure, but the emphasis for those newly-diagnosed is on lifestyle modification ahead of anti-hypertensive drugs. Dr. Sandra Taler presented the rationale for a 130/80 mmHg goal in people with comorbid hypertension and diabetes, drawing on combined data from SPRINT and ACCORD.

5. Dr. Michael Farkouh discussed best practice lipid management for people with diabetes, highlighting a key role for PCSK9 inhibitors in treating individuals in the highest stratum of CV risk. Amgen’s Repatha (evolocumab) and Sanofi/Regeneron’s Praluent (alirocumab) haven’t quite broken into diabetes treatment algorithms, but Dr. Farkouh sees a clear niche for these more potent lipid-lowering agents – people with comorbid diabetes and hyperlipidemia despite maximally-tolerated statin therapy. At the core of Dr. Farkouh’s remarks was a wake-up call to cardiologists: “Type 2 diabetes is a CV disease. We need to get into the game.” We love this sentiment.

6. Dr. Benjamin Scirica outlined a few key wishes he has for the next wave of diabetes CVOTs (where the first wave was safety from DPP-4s and the second wave was efficacy from SGLT-2s and GLP-1s). This includes trials powered for superiority and trials that look more closely at primary prevention. High on his wish list is a UKPDS-like outcomes study that uses advanced diabetes drug classes to normalize blood glucose in patients very recently-diagnosed with type 2 – we were hooked.

7. With a new analysis of FOURIER results, Dr. Marc Bonaca showed that PCSK9 inhibitor evolocumab (Amgen’s Repatha) significantly reduced risk for major adverse CV events in patients with lower-extremity peripheral artery disease (PAD). This cardioprotection was achieved through aggressive lipid-lowering with evolocumab, which also significantly reduced risk for major adverse limb events (MALE; acute limb ischemia, major amputation, or urgent revascularization).

8. While most large CVOTs use time to first MACE event as a primary endpoint, Dr. Sabina Murphy presented a post-hoc analysis of FOURIER showing that Repatha (Amgen’s PCSK9 inhibitor evolocumab) reduced risk for total CV events in the trial, including both first and recurrent events.

9. Following a theme we’ve noticed in 2017, the presence of diabetes on the AHA exhibit hall floor was greater than any year before (we also picked up on this at ACC 2017 in March, and at ESC 2017 in August). We visited eight booths: Amgen, AstraZeneca, Janssen, Lilly/BI, Merck, Novartis, Novo Nordisk, and Sanofi/Regeneron.

Top Nine Highlights

1. New Findings from CANVAS: Risk Reduction for Three-Point MACE Driven by High-Risk Patients, but SGLT-2 Invokana Shows More Consistent Benefits on Heart Failure and Renal Endpoints Across Primary and Secondary Prevention Cohorts

Stanford’s Dr. Kenneth Mahaffey discussed a subgroup analysis of CANVAS suggesting consistent CV and renal benefits to Invokana (canagliflozin) therapy across primary and secondary prevention cohorts. When these CVOT results were announced at ADA, the diabetes field seemed excited that 34% of patients enrolled (n=3,486 out of 10,142) had no history of CV disease, namely because this presented the first opportunity to evaluate an SGLT-2 inhibitor in primary CV prevention – EMPA-REG OUTCOME for Lilly/BI’s Jardiance (empagliflozin) featured no such lower-risk cohort, as all participants entered the study with established CV disease. We got a first look at these highly-anticipated insights in Dr. Mahaffey’s talk, and the results were simultaneously published online in Circulation.

• Three-point MACE (non-fatal MI, non-fatal stroke, or CV death): Risk reduction for this primary outcome was 18% in the secondary prevention cohort with canagliflozin vs. placebo (HR=0.82, 95% CI: 0.72-0.95). The hazard ratio within the primary prevention cohort was higher, at 0.98, with a confidence interval that spanned unity (95% CI: 0.74-1.30). That said, investigators found a non-significant p-value for interaction of 0.18, and Dr. Mahaffey underscored that there was no statistically significant evidence of heterogeneity of canagliflozin's effect across primary and secondary prevention groups. He also showed how the event rate was more than 2x among secondary prevention patients vs. primary prevention patients, explaining that a higher residual CV risk typically translates to a greater absolute risk reduction with a cardioprotective treatment. As a reminder, the original CANVAS results reported a 14% relative risk reduction for three-point MACE with canagliflozin (HR=0.86, 95% CI: 0.75-0.97, p=0.0158 for superiority).
• Hospitalization for heart failure: Relative risk reduction with Invokana was 32% in the secondary prevention cohort (HR=0.68, 95% CI: 0.51-0.90). The hazard ratio for this endpoint was 0.64 in the primary prevention cohort (95% CI: 0.35-1.15), and there was a highly non-significant p-value for interaction at 0.91. In the overall CANVAS trial, Invokana reduced risk for heart failure hospitalization by a 33% (HR=0.67, 95% CI: 0.52-0.87). Dr. Mahaffey described a 2.5x heart failure event rate in the secondary prevention group vs. the primary prevention group, again highlighting the impact of residual risk factors on absolute risk reduction.
• Renal composite endpoint (renal death, renal replacement therapy, or 40% reduction in eGFR): Canagliflozin was associated with a 41% relative risk reduction in the secondary prevention cohort (HR=0.59, 95% CI: 0.44-0.79), and with a hazard ratio of 0.63 (95% CI: 0.39-1.02) in the primary prevention cohort, compared to a 40% relative risk reduction across the entire study population (HR=0.60, 95% CI: 0.47-0.77). The p-value for interaction was 0.73, again highly non-significant. Dr. Mahaffey pointed to a 1.5-fold renal event rate among higher-risk participants vs. lower-risk participants.
• Safety: There was no imbalance in adverse events between primary and secondary prevention cohorts. Canagliflozin increased risk for lower limb amputations despite baseline CV risk, with a hazard ratio of 2.1 (95% CI: 1.4-3.0) in the secondary prevention group and a hazard ratio of 1.5 (95% CI: 0.7-3.3) in the primary prevention group. As initially reported at ADA, canagliflozin was associated with a nearly two-fold increase in lower-extremity amputations across the full integrated dataset (HR=1.97, 95% CI: 1.41-2.75, p<0.001). Dr. Mahaffey mentioned in a separate conversation with us that this amputation signal remains somewhat of a mystery: “The problem is we don’t have an underlying, unifying biological mechanism for why these amputations are happening.” That said, he also emphasized that overall amputation rates were very low (and the majority, 71%, were minor), suggesting that real-world HCPs will have to be cautious and have a patient’s feet diligently monitored if  the joint patient/provider decision is to use canagliflozin for its glucose-lowering, weight loss, blood pressure-lowering, and cardioprotective effects.
• Oxford’s Dr. Angelyn Bethel provided the discussant on this new analysis, and while she was skeptical that Invokana’s benefits to three-point MACE apply in a primary prevention population, she expressed more confidence that the drug’s heart failure and renal benefits do translate. The p-values for interaction on these latter endpoints were more highly non-significant (0.91 and 0.73, respectively) vs. the p-value for interaction on three-point MACE (0.18). Moreover, Dr. Bethel speculated that none of the mechanisms proposed thus far for an SGLT-2 inhibitor’s MACE benefit would extend to those without prior CV events. These hypotheses include changes to volume status impacting heart failure outcomes (which would help prevent heart failure in low-risk individuals, but not necessarily MACE events) as well as an increase in ketone metabolism (this is more efficient for a compromised heart, but those with no history of CV events don’t have a compromised heart per se). During our interview with Dr. Mahaffey, he acknowledged this as a possibility, but still defended the use of Invokana across a broader spectrum of the type 2 diabetes population. As he put it (which we found extremely compelling), “if you’re only going to use this medicine to treat a secondary prevention group, you’re missing out on the important reductions in heart failure hospitalization and renal complications that exist even in the primary prevention group.”
• Very notably, CANVAS was not powered to show superiority within either subgroup. Given the lower event rates (for MACE, heart failure, and renal endpoints) within the primary prevention population, the possibility remains that longer-term follow-up might have accrued a sufficient number of events to show statistically significant benefit. To this end, Dr. Mahaffey pointed to upcoming readouts from CREDENCE (J&J’s outcomes trial for Invokana in diabetic kidney disease) and DECLARE (AZ’s CVOT for SGLT-2 inhibitor Farxiga, which has enrolled >17,000 participants including an even larger primary prevention cohort, and which has a longer median follow-up time of five years).
• According to Dr. Mahaffey, positive CVOTs like CANVAS make the case for more integrated diabetes care teams involving an endocrinologist, cardiologist, PCP, internist, etc. He highlighted a “phenomenal opportunity” for diabetes care to improve in a big way, with an emphasis on outcomes instead of biomarkers, and with therapies in tow that can actually prevent heart attacks, strokes, heart failure, and kidney complications. While high-risk patients stand to benefit the most from cardioprotective, renal-protective agents, down the line, we will hopefully start to see earlier intervention with SGLT-2 inhibitors. The composite benefits to this class cannot be understated – in addition to CV and renal risk reduction, they are convenient oral tablets that are easy to prescribe, with no titration necessary.

2. New Findings from EMPA-REG OUTCOME: Consistent Benefits to Jardiance Regardless of Peripheral Arterial Disease (PAD); Another Reminder that this CVOT Found No Signal for Amputations

Dr. Subodh Verma presented a new post-hoc analysis of EMPA-REG OUTCOME (for Lilly/BI’s SGLT-2 inhibitor Jardiance) that focused on patients with baseline peripheral arterial disease (PAD). Safety and efficacy findings in this subgroup were comparable to the overall CVOT results. P-values for interaction with baseline PAD status were 0.6684 for CV death, 0.5652 for all-cause death, 0.9052 for three-point MACE (non-fatal MI, non-fatal stroke, or CV death), 0.5315 for heart failure hospitalization, 0.9948 for the composite of CV death/heart failure hospitalization, and 0.3282 for incident or worsening nephropathy – note these are all non-significant. As expected, individuals with PAD experienced greater absolute risk reduction for these adverse outcomes, given their higher risk to begin with. For example, absolute risk reduction for CV death was 3.0% in the PAD cohort vs. 2.0% in the non-PAD cohort. Absolute risk reduction for heart failure hospitalization was 2.2% and 1.2%, respectively. After reviewing the efficacy data, Dr. Verma turned to safety. He acknowledged that amputations are of interest when it comes to SGLT-2 inhibitor therapy, since J&J’s Invokana showed a nearly two-fold increase in risk for lower-extremity amputations in the CANVAS trial. No such signal was seen in the initial EMPA-REG OUTCOME analysis, and this held true for people with baseline PAD as well, even though this is a well-known risk factor for amputations (p-value for interaction=0.2752). In fact, the hazard ratio for lower limb amputations among participants with PAD actually trended in favor of empagliflozin (HR=0.84, 95% CI: 0.54-1.32), but it’s important to keep in mind that this represents only a small number of events.

• Dr. Renato Lopes (Duke University, Durham, NC) provided the discussant, and cautioned against over-comparison between CANVAS and EMPA-REG OUTCOME when it comes to the amputation data. He emphasized that amputation was not formally adjudicated in either trial (this is not a classic hard endpoint, and is left up to the joint decision-making of patient/provider). He highlighted differences in study design and study populations, also reminding the room that “one can never rule out the play of chance.” Dr. Lopes confirmed our sense that these two CVOTs on Invokana and Jardiance were very different studies, and we look forward to further analyses that may elucidate the amputation risk with canagliflozin, attributing it to factors other than the molecule or bringing to light criteria that could be used to inform proper patient selection.
• Dr. Lopes described a substantial overlap between diabetes at PAD. According to one of his introductory slides, around 40% of participants in recent PAD trials have diabetes, while 20% of patients enrolled in diabetes trials have PAD. Since CANVAS reported, thought leaders have advocated for more dedicated study into best practice diabetes management for individuals with PAD, because diabetes and PAD compound as risk factors for amputation (read thoughts from Dr. Kittie Wyne in our AADE 2017 report).
• At the start of this session, the new EMPA-REG OUTCOME data was published online in Circulation and Lilly sent out a press release announcing the new positive findings.
• In an interview with Dr. Thomas Seck, VP of of Clinical Developent and Medical Affairs at BI, we learned that PAD affects one in three type 2 diabetes patients over the age of 50. Dr. Seck named PAD as one of the most common CV complications of diabetes, underscoring the importance of addressing this comorbidity in approaches to diabetes management. How does Jardiance fit in? According to Dr. Seck, this sub-analysis provides additional robust evidence for empagliflozin's beneficial effects in a diverse array of diabetes patients.

3. Dr. Mentz Presents First Post-Hoc of EXSCEL: AZ’s GLP-1 Bydureon (Exenatide) Shows Consistent Effects on CV Risk and All-Cause Death Across Risk Quintiles; Questions Loom Large over Primary vs. Secondary Prevention Cohorts in Diabetes CVOTs

In the first post-hoc to-date of AZ’s EXSCEL CVOT, GLP-1 agonist Bydureon (exenatide once-weekly) demonstrated consistent effects on all-cause mortality and three-point MACE regardless of baseline risk score. This data was presented by Duke University’s Dr. Robert Mentz, who reviewed methods for stratifying the study population into risk quintiles. After dividing EXSCEL participants into these five subgroups, the analysis found no significant interaction between risk score and treatment effect, though exenatide was favored by all hazard ratios. For all-cause mortality, hazard ratios ranged from 0.633-0.906 with no directional trend. This means exenatide showed a similar, non-significant relative risk reduction vs. placebo in each quintile, despite a marked increase from 51 events in the lowest-risk group to 527 events in the highest-risk group. A similar pattern was seen for three-point MACE (non-fatal MI, non-fatal stroke, and CV death). Dr. Mentz highlighted non-significant p-values for interaction – 0.2 for all-cause mortality and 0.79 for the primary three-point MACE outcome. The implications of this post-hoc analysis are mixed. On the one hand, exenatide showed a trend toward benefit on hard outcomes that really matter to patients (CV complications and death), and this applied regardless of an individual’s risk score at the start of treatment. On the other hand, after primary EXSCEL results were announced at EASD 2017, showing Bydureon’s narrow miss for CV efficacy (HR=0.91, 95% CI: 0.83-1.00, p=0.06 for superiority vs. placebo), some hypothesized that the neutral result may be due to inclusion of a larger primary prevention cohort, implying that high-risk patients should have derived greater, perhaps statistically significant CV benefit – this was not found according to Dr. Mentz’s presentation. Of course, this is all speculation for now, because we imagine there is much more to unpack here. We eagerly anticipate many more post-hoc analyses of EXSCEL to come at future scientific meetings, which will hopefully contribute further to our understanding of exenatide’s impact on outcomes across a variety of clinical scenarios.

4. 130 is the New 140 in ACC/AHA Blood Pressure Guidelines; Diabetes-Specific Target Strengthened at 130/80 mmHg

The release of new blood pressure guidelines from ACC/AHA generated quite a lot of buzz around the conference center on Monday: With the threshold for hypertension moved down from 140 mmHg systolic to 130 mmHg systolic, nearly half the US adult population could now be diagnosed with high blood pressure. Dr. Sandra Taler presented the rationale for a 130/80 mmHg goal in people with comorbid hypertension and diabetes. This particular recommendation appeared in the most recent NIH Joint National Committee 7 guideline from 2003 as well, although it was not supported by trial evidence at that time. The language around a “lower systolic target, such as <130 mmHg” as “appropriate for certain individuals” has been strengthened in the 2017 guidelines, to “in adults with diabetes mellitus and hypertension, anti-hypertensive drug treatment should be initiated at a blood pressure of 130/80 mmHg or higher with a treatment goal of <130/80 mmHg.” Dr. Taler explained that this evidence-based recommendation accounts for both SPRINT and ACCORD, two seemingly-contradictory studies on the benefits to lower blood pressure. Though ACCORD found no significant CV risk reduction with intensive blood pressure control, Dr. Taler noted that this study was under-powered (indeed, we’ve heard this opinion from many diabetes thought leaders). She displayed combined data from SPRINT and ACCORD to show statistically significant risk reduction for non-fatal MI, stroke, heart failure, and composite CV endpoints. She also discussed a recent meta-analysis of 42 blood pressure trials, 30 enrolling patients with diabetes, which found added CV benefits to more intensive blood pressure-lowering. The new ACC/AHA document recognizes that data at the intersection of hypertension/diabetes is still limited. No RCTs have explicitly demonstrated the benefits of blood pressure <140 mmHg vs. <130 mmHg on clinical outcomes in diabetes. That said, the guidelines have been developed through systematic reviews of the high-quality studies that are out there. Ultimately, we believe the stronger recommendation of <130/80 mmHg could be helpful in encouraging real-world HCPs to intervene more swiftly with more aggressive, comprehensive treatment in diabetes care (targeting not only glucose, but also blood pressure and lipid levels), given that clinical inertia otherwise persists. There is growing consensus in the field that best practice diabetes management addresses these three surrogate markers simultaneously – glucose, blood pressure, and lipids – to help patients avoid microvascular and macrovascular complications. In our view, the new ACC/AHA guidelines are a step in the right direction.

• Importantly, these are guidelines written by a professional cardiology society, not one focused explicitly on diabetes. The recommended DASH diet for people with hypertension, for example, may not be ideal for people with diabetes, and other features embedded within these treatment algorithms will also have to be personalized for patients with multiple comorbidities.
• The updated classification scheme defines “normal blood pressure” as <120/80 mmHg, “elevated blood pressure” as 120-129/80 mmHg, “high blood pressure stage 1” as 130-130/80-89 mmHg, and “high blood pressure stage 2” as >140/90 mmHg.
• There is a major emphasis on lifestyle modification strategies in the new ACC/AHA guidelines. While it’s true that many more people can now be diagnosed with high blood pressure, given the lower threshold, the committee suggests lifestyle prescriptions ahead of prescriptions for anti-hypertensive medications. According to Dr. Paul Whelton, chair of the guideline-writing committee, 14% more adults in the US will fit the diagnostic criteria for hypertension per the revised system, but only one in five of these individuals will require anti-hypertensive drugs. The large majority (80%) should be supported by their HCPs in implementing lifestyle change to get their blood pressure (and other CV risk factors) under control. The implications of this are huge, and we’d love to see a health economic analysis of cost-savings with a lower blood pressure target that motivates more people into action to prevent CV disease.

5. A Wake-Up Call for Cardiologists: “Type 2 Diabetes is a CV Disease. We Need to Get into the Game.” – Dr. Michael Farkouh

Dr. Michael Farkouh (University of Toronto, Canada) discussed best practice lipid management for people with diabetes, highlighting a key role for PCSK9 inhibitors in treating individuals in the highest stratum of CV risk. He credited the diabetes field for mobilizing toward a more comprehensive approach to treatment that considers LDL cholesterol and blood pressure alongside A1c. The ADA recommends at least moderate-intensity statins for people with type 2 diabetes >40 years-old, and suggests ezetimibe for patients who need further lipid control not achieved by statins alone. PCSK9 inhibitors, including Amgen’s Repatha (evolocumab) and Sanofi/Regeneron’s Praluent (alirocumab), haven’t quite broken into diabetes treatment algorithms, but Dr. Farkouh sees a clear niche for these more potent lipid-lowering agents: He argued that they should be considered for people facing very high CV risk, with comorbid diabetes and hyperlipidemia despite maximally-tolerated statin therapy. The benefits to lipid-lowering in diabetes are well-documented. Dr. Farkouh summarized CVOTs on statins, showing a consistent ~30% risk reduction for major adverse CV events overall, which is exaggerated when you look at sub-populations with diabetes, up to a 55% risk reduction. He also pointed to a sub-analysis of FOURIER presented recently at EASD 2017, in which participants with diabetes experienced greater absolute risk reduction for CV events with evolocumab compared to the overall study population (17% relative risk reduction for the primary composite endpoint among people with diabetes vs. 13% across the entire FOURIER trial). This makes sense conceptually, because as Dr. Farkouh reiterated, diabetes is an additional risk factor for CV morbidity and mortality that compounds high LDL in this particular patient population – there’s thus more “room,” in a sense, for more dramatic CV risk reduction. Dr. Farkouh claimed this may be the “sweet spot” for PCSK9 inhibitor therapy, at the intersection of diabetes and dyslipidemia. To be sure, we’re eager to see greater use of PCSK9 inhibitors in diabetes care, because of this growing recognition that lipid control is crucial to achieve the best possible health outcomes for patients. Several thought leaders (spanning the diabetes community as well as the lipid community) have underscored the need for risk stratification in order to gain access to PCSK9 inhibitors for the patients who need them the most (poor reimbursement is the major barrier to uptake for this advanced therapy class currently). Dr. Farkouh presented one compelling risk stratification, and we can only hope that payers are hearing this message, so that people with diabetes/dyslipidemia can benefit from these highly-efficacious products now on the market – it’s a shame to think Repatha and Praluent are available, but still far from accessible.

• At the core of Dr. Farkouh’s remarks was a wake-up call to cardiologists: “Type 2 diabetes is a CV disease. We need to get into the game.” What a fantastically pithy way to capture the paradigm shift happening now in diabetes care – as cardioprotection becomes a central goal in managing hyperglycemia, cardiologists have been officially invited onto diabetes care teams. Moreover, Dr. Farkouh suggested that cardiologists could help accelerate the movement past a glucose-centric view of diabetes, by drawing more attention to hard outcomes. We’re certainly noticing both these themes at AHA 2017: (i) There are many sessions (including this one) dedicated to educating cardiologists – on the implications of high blood glucose for CV outcomes, on the special considerations involved when treating hypertension and hyperlipidemia in patients with diabetes, etc. – so that they feel empowered to treat people with diabetes. And, (ii) there’s an emphasis on outcomes-based approaches to diabetes care, with in-depth discussions of cardioprotective and renal-protective diabetes drugs.

6. Dr. Scirica’s “Wish List” for Future Diabetes CVOTs: Powered for Superiority, Looking at Primary Prevention and Normalizing Blood Glucose in Recently-Diagnosed Patients, Head-to-Head Comparisons

Dr. Benjamin Scirica outlined a few key wishes he has for the next wave of diabetes CVOTs, including trials powered for superiority and trials that look more closely at primary prevention. “Any future study done in this space has to be powered for superiority,” he argued. “Non-inferiority at this point isn’t worth the effort, considering we have other drugs that actually show superiority.” We imagine that limitations on time and/or resources might still lead companies to design CV safety trials in the near-future (after all, a CVOT is a massive investment), but we definitely see Dr. Scirica’s point here, as the bar for new diabetes therapies continues to rise. In order for a new product to be commercially competitive and valuable to patients, demonstrating cardioprotection is becoming increasingly important. This isn’t only our view – ADA Chief Scientific, Medical, and Mission Officer Dr. Will Cefalu expressed a similar opinion in discussing Merck/Pfizer’s SGLT-2 inhibitor candidate ertugliflozin, acknowledging strong phase 3 data but emphasizing that the agent will have to keep pace with others in its class that have already shown significant CV benefit (J&J’s canagliflozin, Lilly/BI’s empagliflozin). Dr. Scirica also posed an all-important question that has emerged through this second wave of positive diabetes CVOTs (where the first wave was safety from DPP-4 inhibitors and from Sanofi’s GLP-1 agonist lixisenatide): Should metformin still be first-line? He implied that the answer is “probably not” for type 2 diabetes patients with established CV disease, but underscored that much more research is needed in primary prevention. Luckily, the diabetes toolkit now features treatment options with greater potency and milder side-effects (especially when it comes to hypoglycemia risk). As such, on Dr. Scirica’s wish list of clinical trials is a UKPDS-like study that uses advanced agents (SGLT-2 inhibitors, GLP-1 agonists) to normalize glucose in newly-diagnosed type 2 diabetes. If we intervene at a much earlier stage of disease development, could we prevent the first instance of a macrovascular complication (rather than the third or fourth, as has been the focus of most diabetes CVOTs to-date)? While we’re not aware of a UKPDS-like trial in the works, J&J management has previously mentioned plans to conduct a CVOT of Invokana in people with prediabetes (we’ve heard no recent updates on this project, and we expect the company is evaluating next steps for canagliflozin clinical development following CANVAS results). We also see an opportunity to look at prediabetes/recent-onset diabetes in the ongoing outcomes trials of SGLT-2 inhibitors in chronic heart failure and kidney disease (EMPEROR, Dapa-HF, and Dapa-CKD enroll participants with and without baseline diabetes). Notably, none of these studies truly represent a primary prevention population for CV disease, because they still enroll people with prior heart failure or otherwise high CV risk, but we’ll be curious to see what insights may be gleaned with regard to normalizing blood glucose. Echoing a similar talk he gave at CMHC last month in his hometown Boston, Dr. Scirica also called for more head-to-head comparisons of cardioprotective therapies in the next wave of diabetes CVOTs. Certainly, a better understanding of how CV effects compare across drugs would be valuable to patients/providers, and would allow for more personalized treatment decisions.

7. PCSK9 Inhibitor Repatha Shows Robust Efficacy in Reducing CV Risk and Adverse Limb Events in People with Peripheral Artery Disease (PAD) – New FOURIER Post-Hoc; Increasing Recognition of the Need to Treat PAD in Context of Hyperlipidemia and Diabetes

With a new analysis of FOURIER results, Dr. Marc Bonaca showed that PCSK9 inhibitor evolocumab (Amgen’s Repatha) significantly reduced risk for major adverse CV events in patients with lower-extremity peripheral artery disease (PAD). This cardioprotection was achieved through aggressive lipid-lowering with evolocumab, which also significantly reduced risk for major adverse limb events (MALE; acute limb ischemia, major amputation, or urgent revascularization). Simultaneous to the start of this session, the post-hoc results were published in Circulation. Among all patients with PAD enrolled in the FOURIER trial (n=3,642 out of 27,564), treatment with evolocumab gave a 21% relative risk reduction for the primary MACE endpoint (CV death, non-fatal MI, non-fatal stroke, hospitalization for unstable angina, or coronary revascularization), with a hazard ratio of 0.79 (95% CI: 0.66-0.94, p=0.0098). For comparison, those without PAD (n=23,922) experienced a 14% risk reduction for this primary composite endpoint with evolocumab vs. placebo (HR=0.86, 95% CI: 0.80-0.93, p=0.0003). While the absolute risk reduction was greater in the PAD subpopulation (as expected, due to higher residual risk at baseline), the p-value for interaction with PAD status was non-significant at 0.40. Risk reduction for this primary endpoint was 15% in the overall FOURIER trial (HR=0.85, 95% CI: 0.79-0.92, p<0.001). For the key secondary outcome of three-point MACE (non-fatal MI, non-fatal stroke, or CV death), evolocumab resulted in a 27% relative and 3.5% absolute risk reduction among people with PAD (HR=0.73, 95% CI: 0.59-0.91, p=0.004) vs. a 19% relative and 1.4% absolute risk reduction among patients without PAD (HR=0.81, 95% CI: 0.73-0.90, p<0.0001). As a reminder, relative risk reduction across the entire study population was 20% for this secondary endpoint (HR=0.80, 95% CI: 0.83-0.88, p<0.001). Dr. Bonaca noted that PAD and prior MI/stroke both confer their own degree of increased risk for MACE. In a pre-specified subgroup of patients with PAD but no prior MI/stroke (n=1,505), evolocumab gave a striking 43% relative risk reduction in three-point MACE (HR=0.57, 95% CI: 0.38-0.88, p=0.0095) as well as a 4.8% absolute risk reduction.

• On major adverse limb events, the analysis found <0.2% risk of MALE in patients with no known PAD, compared to 2.4% risk in patients with known PAD treated with placebo and 1.5% risk in those treated with evolocumab. In patients with PAD and no prior MI/stroke, there was a 57% relative risk reduction for MALE with evolocumab therapy (HR=0.43, 95% CI: 0.19-0.99, p=0.042) and a 1.3% absolute risk reduction. This compares to a 42% relative risk reduction for MALE (HR=0.58, 95% CI: 0.38-0.88, p=0.0093) in the full study population. Dr. Bonaca concluded that use of evolocumab in patients with PAD is safe and highly-efficacious, offering particularly robust reductions in major adverse CV events and limb events, even in those with no prior MI/stroke. When combining MACE and MALE, the number needed to treat in this subgroup to prevent one adverse event was 16 for 2.5 years, with an impressive 6.3% absolute risk reduction. To be sure, the overlap between diabetes and PAD is substantial (43% of PAD patients in this CVOT had a history of diabetes, compared to 36% without PAD). As we imagine greater applications for PCSK9 inhibitors in people with diabetes going forward, we’ll also be excited to see how this potent class of lipid-lowering agents might help those in the highest risk bracket for CV morbidity/mortality, with diabetes, hyperlipidemia, and PAD.
• In a separate conversation with us, Dr. Ransi Somaratne (a cardiologist leading clinical development for Repatha) called these results the most exciting presentation on Repatha at AHA 2017 (this is not a trivial distinction, as at least half a dozen analyses of FOURIER are being presented in Anaheim!). Emphasizing that patients with PAD comprise a very high-risk population, he described how LDL reductions with Repatha resulted in an even greater risk reduction for this subgroup than was seen in the study population as a whole, or in the subgroup without PAD. Dr. Somaratne outlined a couple all-important takeaways for real-world HCPs: (i) He established that this data should encourage providers to pay particular attention to LDL cholesterol in treating individuals with PAD, and he suggested that the findings could be informative in identifying ideal candidates for PCSK9 inhibitor treatment in the real world.

8. New FOURIER Post-Hoc Shows Risk Reduction for Total CV Events (First + Recurrent) with Amgen’s PCSK9 Inhibitor Repatha (Evolocumab)

While most large CVOTs use time to first MACE event as a primary endpoint, Dr. Sabina Murphy presented a post-hoc analysis of FOURIER showing that Repatha (Amgen’s PCSK9 inhibitor evolocumab) reduced risk for total CV events in the trial, including both first and recurrent events. The original results, presented at ACC 2017 in March, analyzed 2,907 instances of the primary composite endpoint (CV death, non-fatal MI, non-fatal stroke, hospitalization for unstable angina, or coronary revascularization). Participants who experienced a non-fatal CV event continued to be followed, and Dr. Murphy shared that there were 1,999 additional events adjudicated during the course of the study. Evolocumab reduced risk for these recurrent events by 26% (risk ratio=0.74, 95% CI: 0.65-0.85), and for total events (first + recurrent) by 18% (risk ratio=0.82, 95% CI: 0.75-0.90, p<0.001). This compares to a 15% risk reduction with evolocumab vs. placebo when looking at time to first occurrence of the primary composite endpoint, as initially reported (HR=0.85, 95% CI: 0.79-0.92, p<0.001). Three-point MACE was a key secondary endpoint in FOURIER, and Dr. Murphy reported a 21% relative risk reduction for recurrent events (risk ratio=0.79, 95% CI:0.61-1.02) as well as a 19% relative risk reduction for total events (risk ratio=0.81, 95% CI: 0.73-0.90, p<0.001). For comparison, evolocumab was associated with a 20% relative risk reduction for time to first occurrence of three-point MACE (HR=0.80, 95% CI: 0.73-0.88, p<0.001). This supports the very robust results from FOURIER, building upon a compelling evidence base for Repatha’s cardioprotective effects (indeed, Amgen is expecting an FDA decision on a CV indication for the drug in early December). During Q&A, Dr. Murphy emphasized that both time to first event analysis and total event analysis offer valuable insight: “When we look at first event, we’re really testing the hypothesis of the trial – does this work? When you look at total events, you’re looking at overall clinical burden of this therapy – how will this ultimately affect the patient? These are two complementary questions, and we think it’s helpful to answer both of them when you’re evaluating new therapies.”

9. Big Diabetes Presence in the Exhibit Hall; Novo Nordisk Debuts First-Ever AHA Booth

Following a theme we’ve noticed in 2017, the presence of diabetes on the AHA exhibit hall floor was greater than any year before (we also picked up on this at ACC 2017 in March, and at ESC 2017 in August). The year started off strong, with Lilly/BI’s SGLT-2 inhibitor Jardiance (empagliflozin) becoming the first diabetes drug with an FDA-approved CV indication in December 2016 – full force rollout of the new label began in 2017. In June, at the ADA’s 2017 Scientific Sessions, another positive diabetes CVOT was presented: J&J’s CANVAS trial for Invokana (canagliflozin) showed that a second SGLT-2 inhibitor also reduces risk for adverse CV outcomes in a clinically-meaningful, statistically significant way. Then, in August, the FDA approved a CV indication for Novo Nordisk’s GLP-1 agonist Victoza (liraglutide). These are just the highlights. Over the course of this year, we’ve heard louder buzz surrounding a paradigm shift in diabetes care, with CV risk mitigation now at the center. This signals an important push toward outcomes-based medicine. It empowers diabetes care providers to treat more than the biomarker of A1c, because they can prescribe medicines that will prevent heart attacks, strokes, and CV death in their patients. It also shines a spotlight on the overlap between diabetes and CV disease, inviting cardiologists onto diabetes care teams. As such, we’re delighted to see a growing presence of diabetes content at cardiology conferences, and this includes the exhibit hall. We visited eight booths: Amgen, AstraZeneca, Janssen, Lilly/BI, Merck, Novartis, Novo Nordisk, and Sanofi/Regeneron. See below for a detailed recap of each.

Exhibit Hall

Amgen

With more than half a dozen new analyses on PCSK9 inhibitor Repatha (evolocumab) being presented at this year’s AHA, Amgen came to Anaheim in full force, and the company’s exhibit hall display was no exception. By our observation, Amgen had the largest (and perhaps most heavily-staffed) booth on the exhibit hall floor, drawing attendees in with multiple ladder toss “CholesterBall” stations where participants aimed to “escape high LDL cholesterol.” The vast majority of signage was dedicated to Repatha, including a bold overhanging banner calling on cardiologists to “help your patients escape high LDL-C.” Another poster highlighted data to show that Repatha dosed every two weeks on top of a statin gave up to 77% additional LDL-lowering. Informational touchscreen panels invited booth visitors to explore patient stories, featuring categories of previous MI, previous stroke, and atherosclerotic CV disease with heterozygous familial hypercholesterolemia (HeFH). A very large TV played a video on loop, reminding people that 80% of patients with atherosclerotic CV disease don’t meet an LDL target of <70 mg/dl. Around the corner, a living-room style setup promoted Amgen’s heart failure drug Corlaner.

Amgen reps reminded us that Repatha has a fairly narrow indication, for the treatment of familial hypercholesterolemia or atherosclerotic CV disease (this may be expanded soon, as an FDA decision on a CV indication is expected by December 2). This may be part of the reason why reimbursement has been a challenge thus far – we’ve heard that prior authorizations are all but necessary, and often difficult to get approved – and why commercial uptake of the PCSK9 inhibitor class (also including Sanofi/Regeneron’s Praluent) has trended below expectations. All this said, Amgen seems as dedicated as ever to the Repatha franchise.

AstraZeneca

The bulk of AZ’s booth was dedicated to antiplatelet therapy Brilinta, though one interior wall focused on once-weekly GLP-1 agonist Bydureon (exenatide), SGLT-2 inhibitor Farxiga (dapagliflozin), and Xigduo (dapagliflozin/metformin fixed-dose combination), with touchscreen panels featuring case studies and patient stories alongside informational brochures. Despite the fact that EXSCEL did not show CV efficacy for Bydureon (exenatide narrowly missed the statistical threshold for superiority, with an upper bound of 1.00 on the 95% CI), AZ still seems very positive about the benefits of Bydureon treatment. The company recently received FDA approval for a new Bydureon autoinjector (branded Bydureon BCise), which eliminates the need for reconstitution and makes the entire injection process much more patient-friendly – we look forward to seeing the autoinjector up close in a future exhibit hall, once it has launched in the US (expected in 1Q18). A small medical information section within AZ’s booth was dedicated to the Dapa-HF, Dapa-CKD, CVD-REAL, and DapaMech (actually three trials) studies on Farxiga. This confirmed our sense that AZ is highly committed to clinical and commercial development of its SGLT-2 franchise, not only for diabetes but for the spectrum of cardio-renal-metabolic syndrome

J&J (Janssen)

Janssen’s diabetes products were unassuming in a back corner of the company’s vast booth, with two small monitors discussing SGLT-2 inhibitor Invokana (canagliflozin) and fixed-dose combination Invokamet (canagliflozin/metformin). CV drug Xarelto (a blood thinner) was the clear focus of Janssen’s exhibit. We anticipate more emphasis on the Invokana franchise if/when the FDA approves a new CV indication for the drug, based on 14% risk reduction for three-point MACE (non-fatal MI, non-fatal stroke, and CV death) in the CANVAS trial. As we understand it, companies are very limited in what they can say with regard to a product’s off-indication benefits until this information is updated in the label. Janssen filed a Supplemental New Drug Application (sNDA) to this end in early October, and an FDA decision on the label update is expected between August-October 2018. If all goes well, the company’s exhibit hall booth may look markedly different at AHA 2018, next November in Chicago.

Lilly/BI

Lilly/BI dedicated a large, spacious booth to SGLT-2 inhibitor Jardiance (empagliflozin), the only “FDA-approved type 2 diabetes pill indicated to reduce the risk of CV death.” Benches surrounded a large rotating heart near the center of the exhibit, and reps invited passersby to learn why providers (including cardiologists) choose Jardiance. Lilly/BI have been refining their booth for a cardiology audience since the Jardiance label update was approved (December 2016), and this time around, we noticed a greater emphasis on the convenient oral dosing of the SGLT-2 inhibitor. Interestingly, Dr. Neil Poulter suggested at ESC 2017 that cardiologists shy away from injectable drugs (they seem more complicated to titrate, and conjure up images of hypoglycemia) – putting aside the issue of whether or not this is reasonable (GLP-1 agonists are cardioprotective without hypo risk), Lilly/BI’s spotlight on oral administration of Jardiance could be strategic for a cardiology meeting. We wonder (and hope) if in a couple years, Lilly’s AHA display will feature GLP-1 agonist Trulicity alongside Jardiance for cardioprotection; the REWIND CVOT is expected to complete in July 2018. For now, we’re so excited to see an increased diabetes presence at AHA, not to mention diabetes drugs being touted for their positive CV effects.

Merck

Merck’s booth featured bright signs highlighting superior reductions in A1c, fasting plasma glucose, and postprandial glucose excursions with DPP-4 inhibitor Januvia (sitagliptin) vs. metformin monotherapy. Right by the company’s classic frozen yogurt stand, visitors could learn all about the incretin effect on an eye-catching visual display: Consuming a meal stimulates the release of incretin hormones GLP-1 and GIP, but the DPP-4 enzyme inactivates these proteins. Enter Januvia, a DPP-4 inhibitor that blocks this action, thereby increasing circulating levels of GLP-1 and GIP. These active incretins then stimulate insulin secretion from the pancreatic beta cells, while suppressing glucagon secretion from the pancreatic alpha cells. We found it notable that Merck was heavily promoting Januvia to an audience of cardiologists, despite the fact that sitagliptin showed strong CV safety but no sign of CV benefit in the TECOS trial. In the age of the CVOT, diabetes thought leaders have invited cardiologists into their care teams, encouraging more collaboration among experts to achieve best practice in diabetes management, which now more than ever emphasizes CV risk reduction. We’d love to see Merck (among others) knock down the barriers that keep cardiologists from treating diabetes, and this certainly starts with sharp education – in this case, education about one of the most highly prescribed branded diabetes drugs, which boasts great safety/tolerability and glucose-lowering efficacy without hypoglycemia risk.

Novartis

Novartis occupied a large booth near the center of the AHA exhibit hall, showcasing heart failure drug Entresto. Bold signs in dark blue and yellow called attention to the “true risk of heart failure” – a single heart failure hospitalization increases an individual’s risk of death six-fold, 50% of patients diagnosed with chronic heart failure live <five years, and death due to heart failure is more common than death attributed to stroke and MI combined. Entresto was presented as a highly-efficacious treatment that could reduce this burden. Interactive iPads were scattered throughout the booth, where conference attendees could learn about outcomes with Entresto, including 20% risk reduction for CV death, 16% risk reduction for all-cause mortality, and 21% risk reduction for heart failure hospitalization. While there was no explicit mention of diabetes, we’re keenly interested in this application of Entresto, because heart failure is a common complication of diabetes, and because of sub-analysis of PARADIGM-HF presented at ACC 2017 showed the drug’s A1c-lowering effect and risk reduction for new initiation of insulin therapy over three years in participants with diabetes at baseline.

Novo Nordisk

Novo Nordisk made its AHA exhibit hall debut with a bustling booth all about GLP-1 agonist Victoza (liraglutide) and its new CV indication. A large overhead banner promoted the concept of “A1CV” – that is, Victoza offers profound A1c reductions as well as meaningful risk reduction for major adverse CV events (including non-fatal MI, non-fatal stroke, and CV death). Another sign within the booth emphasized that Victoza is the only diabetes therapy approved to delay or prevent all components of three-point MACE (whereas Lilly/BI’s SGLT-2 inhibitor Jardiance is approved specifically to reduce CV death). We noticed that this exhibit seemed more vibrant than Novo Nordisk’s booth at ESC 2017 in Barcelona, which makes sense, since this earlier cardiology conference took place in late August, just a couple days after the Victoza label was updated in the US and just a few weeks after the EU label update. Novo Nordisk’s presence on the exhibit hall floor at a cardiology meeting is notable in itself, as this signals a shift in diabetes care toward outcomes-based medicine – in this case, CV outcomes, which introduces a whole new audience in the cardiology community. This is a big deal. For the first time, people with diabetes can take medicine that impacts concrete health outcomes, not just a biomarker (A1c), and it’s crucial that companies like Novo Nordisk follow-up on exciting CVOT data with concerted efforts to educate prescribers. We imagine we’ll be hearing even more about Novo Nordisk’s diabetes portfolio at ESC, AHA, and ACC congresses going forward.

Reps at the Novo Nordisk booth scanned visitor badges and found out which local payers in the individual’s hometown reimburse Victoza completely, or offer the drug at the lowest possible co-pay. As one rep put it, formularies drive prescription habits these days – the company is aware of this, and is committed to making its cardioprotective therapy as accessible as possible. We liked this very practical, hands-on step, helping cardiologists learn more about Victoza coverage in their area.

Sanofi/Regeneron

A sizable Sanofi/Regeneron display focused mainly on PCSK9 inhibitor Praluent (alirocumab), but we were excited to see a corner dedicated to fixed-ratio combination Soliqua (insulin glargine/lixisenatide), next-gen basal insulin Toujeo (insulin glargine U300), and even mealtime insulin Apidra (insulin glulisine). While Soliqua has yet to take off commercially, following its US launch in early 2017, Sanofi management has expressed high hopes for this highly-efficacious product in the real world. Company reps presented Soliqua as an intensification option for patients who aren’t meeting targets with Lantus (insulin glargine U100), while letting them feel as if they’re taking the same amount of medication. Indeed, Soliqua is only indicated in the US for patients already on one of its components, which we see as a possible limiting factor to uptake. We happily noted in Sanofi’s booth that Soliqua has a formulary-independent $0 co-pay card for individuals with commercial insurance.

Paralleling what we heard in Amgen’s booth on PCSK9 inhibitor Repatha (alirocumab), Sanofi reps reinforced Praluent’s somewhat limited indication for patients with either atherosclerotic CV disease or heterozygous familial hypercholesterolemia who need additional LDL-lowering. One rep told us that providers visiting the booth are most curious about who they can give Praluent to, and they’re focused on targeting the right patient. We’ve heard this is key in getting prior authorizations approved for the drug – documenting that a patient fits the Praluent indication and has already tried maximally-tolerated statin therapy. Eye-catching signs within the booth promoted a once-monthly 300 mg dose of Praluent (lower injection burden), given as two simultaneous injections.

-- by Ann Carracher, Payal Marathe, and Kelly Close