by Steven Breazzano, Ning Yang, Ph.D., and Andrew McDonald
Data Comparison -To demonstrate efficacy, the FDA wants to see statistically significant mean weight loss vs. placebo (mean difference of 5% or more), and categorical representations of how many patients achieve 5% or more weight loss (proportion of subjects who lose 5% should be more than 35%, and roughly double the placebo group). The FDA further believes that improvements in blood pressure, lipids, glycemia, or other areas commensurate with the degree of weight lost are important indicators of an effective weight-management product. Also, changes in common weight-related comorbidities should be documented and considered. Patients with type 2 diabetes are especially difficult to treat and have unique safety concerns, thus leading to the obesity companies running separate trials for these patients or categorizing these patients. Overall, it is important to remember that while there are risks associated with taking medications for obesity, patients who do not respond to treatment after a few months will likely discontinue the drug. This discontinuation recommendation was included in the label for Arena's BELVIQ and Vivus' Qsymia. In addition, how applicable these controlled, clinical trials are to the real world is an open question. Will patients modify their diet and increase physical activity away from the controlled clinical trial setting? Or will they maintain their current lifestyle and hope an obesity drug takes care of everything? As might be expected, all the trials show significant placebo response, as patients were counseled on traditional weight loss methods. Even though each company ran their own trials, comparisons are relevant due to the similarity in certain trials with patient numbers, baseline characteristics, duration, and endpoints.
|Combined Phase 3||OB-302 (Equip)||OB-303 (Conquer)||NB-301||NB-303|
|Patients Enrolled (Phase 3 Trials)||7190||1267||2487||1742||1496|
|Mean weight or BMI||~100 kg, 36||37||36.6||100 kg, 36|
|Dropout rate in placebo arm (%)||51%||53%||43%||50.10%||46.30%|
|Dropout rate in active arm (%)||44%||top dose 41%||mid-dose 31%, top-dose 36%||49.20%||46.30%|
|Weight Loss at 52 (or 56) Weeks (% of baseline, placebo)||-2.50%||-2.10%||-1.60%||-1.30%||-1.20%|
|Weight Loss at 52 (or 56) Weeks (% of baseline, drug)||-5.80%||high-dose -11%||mid dose -8%, high dose -10%||-6.10%||-6.40%|
|Difference, placebo adjusted||-3.30%||-9%||mid-dose -6%, high dose - 8%||-4.80%||-5.20%|
|Patients losing 5% or more of weight, placebo||22.60%||17%||21%||16.40%||17.10%|
|Patients losing 5% or more of weight, active||47.00%||high-dose 67%||mid-dose 62%, high-dose 70%||48%||50.50%|
|Difference, placebo adjusted||24.50%||high-dose 50%||mid-dose 41%, high-dose 49%||31.60%||33.40%|
|Patients losing 10% or more of weight, placebo||8.70%||7.40%||7.40%||7%||6%|
|Patients losing 10% or more of weight, active||22.40%||high-dose 47.2%||mid-dose 37.3%, high-dose 47.6%||25%||28%|
|Difference, placebo adjusted||13.80%||high-dose 39.8%||mid-dose 29.9%, high dose 40.2%||18%||22%|
|Notes: all above data is ITT-LOCF (intention-to-treat, last-observation-carried-forward)|
|ARNA - Does not include 2nd year of trial or Type 2 dm study, 52 week endpoint|
|VVUS - Does not include followup study (Sequel), and data is for the mid-dose except where indicated, study endpoint 56 weeks|
|OREX - 56 week endpoints, NB-303 presented separately due to re-randomization at week 28, NB 32 dose|
|Efficay Sumary - Type 2 DM||ARNA||VVUS||OREX|
|BLOOM DM||OB-302 (Equip)||NB-304|
|Patients Enrolled (Phase 3 Trials)||499||1267||505|
|Mean weight or BMI||103.5 kg||42||~106 kg|
|Dropout rate in placebo arm (%)||38%||53%||41.20%|
|Dropout rate in active arm (%)||34%||top dose 41%||47.80%|
|Weight Loss at 52 (or 56) Weeks (% of baseline, placebo)||-1.60%||-2.10%||-1.80%|
|Weight Loss at 52 (or 56) Weeks (% of baseline, drug)||-4.70%||high-dose -11%||-5.00%|
|Difference, placebo adjusted||-3.10%||-9%||-3.80%|
|Patients losing 5% or more of weight, placebo||16.10%||17%||18.90%|
|Patients losing 5% or more of weight, active||37.50%||high-dose 67%||45%|
|Difference, placebo adjusted||21.30%||high-dose 50%||26.10%|
|Patients losing 10% or more of weight, placebo||4.40%||7.40%||5.70%|
|Patients losing 10% or more of weight, active||16.30%||high-dose 47.2%||18.50%|
|Difference, placebo adjusted||11.90%||high-dose 39.8%||13.00%|
|Notes: all above data is ITT-LOCF (intention-to-treat, last-observation-carried-forward)|
|VVUS - all Equip results|
ARNA - The efficacy of Arena's BELVIQ was evaluated in 3 large trials: BLOOM, BLOSSOM, and BLOOM DM (type 2 diabetes). Of the three major candidates, Arena's drug resulted in the least amount of weight loss. Looking at the data in the efficacy summary, the patients in the large phase 3 trials (BLOOM and BLOSSOM, combined in the labeling information) showed statistically significant weight loss relative to placebo after 1 year, and the number of patients losing >5% and >10% of their body weight was statistically significant. However, the year 1 placebo-adjusted weight loss achieved in patients treated with BELVIQ was only 3.3 kg, or 7.2 lbs (with a modified intent to treat analysis). A substantial percentage of randomized subjects withdrew from each study prior to week 52. Although not shown in the above table, patients in the second year of treatment put back on weight, but remained below their original levels.
BLOOM-DM (study 3, 604 adult patients, inadequately controlled type 2 diabetes) showed more modest weight loss, but given the trial enrolled the tough category of patients with type 2 diabetes, this is expected. Also, improvements in other indicators in the BLOOM-DM trial were impressive and statistically significant. Side effects included nausea and toothache, and fatigue.
VVUS -The efficacy of Vivus's Qnexa was evaluated in 3 large trials: Conquer, Equip, and Sequel. The CONQUER study included 2,487 overweight and obese patients (1,737 females and 750 males) with high blood pressure, high cholesterol, or type 2 diabetes. The average baseline BMI of the study population was 36.6 and baseline weight was 227 pounds. Patients had a 4-week dose titration period followed by 52 weeks of treatment. The study was a randomized, double-blind, placebo-controlled, 3-arm, prospective trial with patients randomized to receive once-a-day treatment with mid-dose Qnexa, full-dose Qnexa or placebo. Patients were asked to follow a hypocaloric diet representing a 500-calorie/day deficit and advised to implement a simple lifestyle modification program. There was a low dropout rate in the trials due to adverse events (overall, 69% completed the trials vs. low 50s for placebo), and patients reported increased side effects of dry mouth, tingling, and altered taste in slightly higher rates on Qnexa vs. placebo.
The EQUIP study included 1,267 morbidly obese patients (1,050 females and 217 males) across 93 centers in the United States. The average baseline BMI of the study population was 42.1 and baseline weight was 256 pounds. Patients were also asked to follow the same hypocaloric diet described for Conquer. Among patients who completed the top dose course of treatment, 83.5% lost ≥5%; 67.7% lost ≥10%; and 48.1% lost ≥15% of their baseline weight; and 45%, 19%, and 7%, respectively for the low dose.
In Sequel, a 1 year extension trial of Qnexa that enrolled 676 overweight and obese patients, weight loss was relatively durable in the mid-dose, which is the most likely the main therapeutic dose for patients. Average weight loss at week 108 (including the first year on CONQUER) was 9.3% and 10.5%, respectively, for the mid and top dose of Qnexa as compared to 1.8% for the placebo group. The completion rate in SEQUEL was approximately 83% for both Qnexa doses and 86% for the placebo group. Importantly, placebo patients had a three times greater likelihood to progress to type 2 diabetes compared to subjects receiving top dose Qnexa and a two times greater likelihood than patients on mid dose Qnexa. Discontinuations due to adverse events were 4.5% and 4.4% for the mid and top dose, respectively, and 3.1% for the placebo group. While VVUS did not run a separate trial for patients with type 2 diabetes, approximately 20% of SEQUEL subjects had type 2 diabetes at baseline and had been treated with lifestyle modifications alone or single-agent metformin. In this subgroup, weight loss associated with Qnexa had a favorable impact on glycemic control without a need for added oral hypoglycemic agents. In subjects without type 2 diabetes at baseline, the favorable effects of weight loss on insulin sensitivity and glycemia were associated with decreased progression to type 2 diabetes during the 2-year course of the study.
OREX -Of the three major candidates, Orexigen's results fall in the middle with respect to efficacy, although the efficacy more closely resembles Arena's. Orexigen evaluated Contrave in four Phase 3 clinical trials, and all the trials met their co-primary endpoints as well as key secondary endpoints. Contrave showed a significant reduction in body weight, improvements in cardiovascular and metabolic risk factors, and reductions in selected food craving measures. In these four approximately year long, randomized, double blinded, placebo-controlled trials, or COR, program, Contrave was generally well tolerated by patients, with an overall safety profile that was consistent with its individual components. Analogous to its competitors, many patients did not complete the clinical trials, and dropout rates were high in both the placebo and active arms, with 40-50% of patients dropping out.
COR-I (NB-301) was a 58-week study assessing the safety and efficacy of Contrave in 1,742 non-diabetic, obese patients. Like ARNA and VVUS, patients were also counseled with diet and exercise. COR-II (NB-303) was a similar study, and was conducted over 56-weeks and designed to assess the safety and efficacy of Contrave in 1,496 healthy, non-diabetic, obese patients as well.
COR-Diabetes (NB-304) was a 56-week study designed to assess the safety and efficacy of Contrave in 505 obese subjects who also have been diagnosed with Type 2 diabetes. 45% of patients on Contrave in this trial lost 5% of their body weight, compared to 19% on placebo. 18% of patients on Contrave in this trial lost more than 10% of their body weight, compared to 6% on placebo. These results were statistically significant.
COR-BMOD (NB-302) was a 56-week study designed to evaluate the safety and efficacy of Contrave alone or when combined with intense diet, exercise and behavior modification in approximately 800 patients. Seemingly due to the intense nature of the exercise and behavior modification components of the clinical trial, the patients on placebo responded well.
As shown by the data, efficacy is modest for and the clinical dropout rates are high. During the trials, a slight increase in heart rate was also noted, and this is possibly why the FDA is requiring Orexigen to rule out an increased risk of cardiovascular events prior to approval.
Overall, in terms of absolute weight loss, Qsymia is the most efficacious of the three. However, from a safety standpoint there is the issue of birth defects, and the increase in resting heart-rate. As the label states, the clinical significance of the increase in heart rate is not clear. One additional point to keep in mind is that Arena's BELVIQ (lorcasarin) has not yet been evaluated in combination with phentermine, which was one component of fen-phen. This new combination could lead to pronounced efficacy for BELVIQ.
Market Sizing and Considerations - Obesity is clearly a multi-billion dollar market opportunity with an estimated 1/3rd of the US population obese and growing rapidly. To date however, no obesity drug has hit blockbuster status, which may be defined as greater than $1 BB in annual sales. At fen-phen's 1996 peak though, an estimated 14-18 MM prescriptions were written, and sales were approximately $173 MM. In 2009, the anti-obesity drugs Xenical, Alli (Xenical over the counter), and Meridia (now withdrawn from most markets) each had worldwide annual sales of $300 MM to $350 MM. Generic products, including phentermine, generated $300M to $600M in sales. Currently phentermine is written as a short-term obesity prescription, but it is not indicated for long-term use. Previously, worldwide Xenical sales reached nearly $750 MM in 1999, when Xenical was first available in the US. Since then, sales have tapered off. However, increasing physician and patient realization of the risk factors associated with obesity, combined with both the increased prevalence and incidence of obesity, bode well for market penetration and uptake of Qsymia and BELVIQ. Unlike many drugs that require a few years to ramp up sales, previous obesity drugs have seen very quick launches followed by a drop-off in enthusiasm.
The high clinical dropout rate seen in Phase III trials could dampen optimism and foreshadow future market penetration. To achieve blockbuster status, new patient starts must outpace treatment discontinuation. This phenomenon has plagued obesity drugs in the past and may be responsible for the lack of a blockbuster drug in the space. In ARNA's pivotal phase 3 trials, discontinuation rates hovered around 50% in both arms. Competitor VVUS fared significantly better, with only 31% discontinuing at the mid-level dose that most patients will take. Interestingly, and quite consistently, most of the discontinuations were not due to treatment related side effects. This suggests that efficacy will be a primary driver for patients to continue with their medications. Based on the similarity in efficacy between BELVIQ and Xenical, ARNA peak US sales may be limited to $1 BB annually or less, much of which may come in the first few years as enthusiasm drives patient uptake. Although no data on pricing has been released, this forecast crudely represents 1 MM patients taking the drug for a year at a price of $1000/year, or $83/month; not unreasonable for an out of pocket expense in the first few years.
Furthermore, despite Xenical's well-publicized (and sometimes embarrassing!) side effects, discontinuations due to side effects were not common in the trials, validating the comparison. As mentioned earlier, VVUS notably had far less discontinuations and superior efficacy, which bodes well for market penetration. Due to the REMS in place however, patient uptake may be slow for Qsymia. In the next few years VVUS could have a large share of a potentially $2BB US market 4 years from now, which represents 2 million patients per year paying $1000/year on an obesity treatment, a small percentage of the overall obese population (~2% penetration of 100 MM obese patients). In addition, VVUS, and potentially OREX, may face competition from the individual generic components that are cheaply available. However, formulation and dose considerations, along with a strong REMS, encourages use of the branded product.
While Arena did not provide any guidance on the launch other than to say Eisai will be focusing its primary care sales force on it, Vivus has been discussing its plans for the launch. Vivus anticipates that initially Qsymia will be cash pay without good coverage, but they hope to improve coverage by the end of the year. Vivus' data that demonstrated a threefold reduction in progression to type 2 diabetes, a very expensive indication for insurance companies, strengthens the argument for reimbursement of obesity treatments. However, current data from the UK suggests that insurance reimbursement is not a dominant factor determining market penetration. Currently in the UK these costs are reimbursed, and market penetration remains low.
Valuation - With market capitalizations and EV's of nearly $2 BB and $1.5 BB for VVUS and ARNA respectively, expectations for the launches are quite high. Given that peak sales for the space in the US in a few (4) years may reach $2BB, with VVUS capturing 50%, or $1 BB of annual sales, applying a 4x revenue multiple to reflect the wholly-owned asset and discounting back 4 years at 15% yields an implied market capitalization of $2.3 BB, and a share price of roughly $22 (similar to where the stock trades now), based simply on US sales. However, significant costs should be incurred along the way (CVOT trial, etc.) and there are significant commercialization and execution risks associated with launching a drug with a small sales force. However, EU partnerships and partnering the ED drug represent upside potential. Assuming ARNA (and Eisai) receives the other 25% of the future $2 BB market, or $500 MM in annual revenues, would imply roughly $180 MM in annual income (plus purchase price adjustments) flowing to ARNA. Because this revenue stream takes into account distribution, marketing and a significant portion of post-marketing studies, most of this income will fall to the bottom line. Performing a similar analysis on ARNA, and applying 4x revenue multiple followed by discounting back at 15% for 4 years yields a crude expected market capitalization of approximately $1.1-1.2 BB and a share price of roughly $7-8 (close to current valuation), but again most of this will fall to the bottom line and Arena will not incur nearly as many costs as Vivus.
Orexigen, featuring the smallest enterprise value (~$360 MM) and market capitalization (~$500 MM) of the three, is more speculative and will not launch its drug for at least 2 years. Given the similar drug efficacy to ARNA, assuming Contrave delivers $500 MM in annual revenues (25% of the market), double-digit royalties as well as a co-promote option may deliver $100 to 200 MM to the bottom line. Again applying a 4x revenue multiple to $500 MM of sales and discounting back 6 years at 20% to reflect the greater regulatory risk and late mover implies a market capitalization of ~650-700 MM, or roughly $6-7/share, slightly above the current share price. Furthermore, the fact the drug has not cleared the FDA and requires a positive interim analysis remain significant risk factors. Overall, it is reasonable to believe physicians will rotate and "try-out" different therapies for patients. As these crude valuations imply, the market appears to be expecting significant sales in the obesity space to justify these market capitalizations and share prices.
Conclusions - Obesity is a significant unmet need and represents a huge burden on the healthcare system. Now with two FDA approved treatments preparing for launch, and potentially another on the way, significant commercialization and execution risks loom. Taking a look at individual companies reveals pros and cons for each.
With Arena, positives include the fact the drug is already approved and they already have a marketing partner (Eisai) with respectable terms. Of the three companies, BELVIQ is a new chemical entity (NCE) and possess the strongest patent estate. On the other hand, the efficacy is modest and the high dropout rate in the clinical trials may limit potential patient uptake and future peak sales.
Vivus also recently received approval, which removes regulatory risk, and also has the best efficacy of the three, boding well for achieving blockbuster status. In addition, the entire asset is solely owned by Vivus, and the current balance sheet is strong. However, launching a drug without a marketing partner may be difficult and the REMS may hinder patient uptake. Furthermore, market share may be eroded by generic substitutes and the intellectual property position may pose an issue down the road. Also, expensive post approval studies are required. Orexigen still has regulatory risk, but has already received a positive AdCom vote, they have a marketing partner (Takeda) which will also cover post-approval studies, and the CVOT has good odds of succeeding. The drug's efficacy falls in the middle, but will not be on the market for at least two years. Like Vivus, generic substitution is also a potential liability.