INTRODUCTION

With patent expirations and challenges in small molecule research and development (R&D), many of the drugs in the R&D pipeline are biologics, with therapeutic indications as diverse as oncology, inflammatory conditions, multiple sclerosis, orphan diseases and other conditions.

Drugs for both cancer and orphan conditions are typically very high cost, leading to a relatively small number of patients having much higher per patient drug costs compared to average per patient drug costs. The costs of specialty drugs have been rising at double-digit rates over the last several years; this is expected to continue (Figure 1 see end notes for definition of specialty drug).

Figure 1
figure 1

 Specialty drug trend1: 2009–2012. Source: Express Scripts 2009 Drug Trend Report, p. 95.

Full size image

Since biotechnology agents have a high per patient cost and are already a key driver of total drug spending, payers have developed a number of strategies to manage their utilization, particularly for self-administered2 (oral and subcutaneously administered) agents. Another reason that management strategies for these types of agents have developed is that they are typically covered under pharmacy benefits, which have more of a cost control focus, and better data upon which reimbursement strategies can be justified. Strategies to manage drugs that are infused, and frequently covered under the medical benefit, are evolving, in part because of the increased role of pharmacy departments and Specialty Pharmacy Programs (SPPs) in management of these patients. Notably, the Centers for Medicare and Medicaid Services (CMS) opened a National Coverage Decision (NCD) for Provenge, a vaccine for patients with advanced prostate cancer, shortly after Provenge was launched.

The total enrollment in Medicare (including Medicare Part D, the Medicare Prescription Drug program, which covers self-administered drugs for Medicare beneficiaries) and state Medicaid programs is growing. The power of managed care is also increasing, because of its role in commercial insurance (typically employer-based), as well as Medicare Part D and Medicare Advantage (Part C, which covers medical and pharmacy benefits for Medicare beneficiaries). The Patient Protection and Affordable Care Act (PPACA) is likely to substantially expand managed care's role, both because of the expansion of Medicaid (which, particularly for younger beneficiaries, typically works with managed care) and managed care's likely role in the new state-based health insurance exchanges which start in 2014.

At the same time, the importance of health-care cost control is increasing, due to the multi-year recession, expansions in coverage, and continued unsustainable growth in health-care costs. Even before recent coverage expansions and the recession, the Congressional Budget Office estimated total health-care spending rising from 16 per cent of Gross Domestic Product in 2007 to 25 per cent in 2025.3

IMPLICATIONS FOR THE DEVELOPMENT OF BIOTECHNOLOGY AGENTS

Biologics that are in early development now, and launched in the 2012 to 2015 period, are likely to face a greater level of payer scrutiny, and more restrictions on access, than has occurred to date. Oncology drugs, and drugs administered incident to health-care professional visits and covered under fee-for-service Medicare Part B may well have fewer restrictions than other biotechnology agents, but we expect that the payer environment for these drugs may well change as well.

Thus, there will be four hurdles to a drug's success (Figure 2). Compound identification is the first hurdle, conducting the clinical trial program and receiving FDA approval is the second hurdle; both hurdles are commonly recognized and understood. The third hurdle is physician acceptance and prescribing; the fourth hurdle is payer coverage and reimbursement (which of course affects physician prescribing as well).

Figure 2
figure 2

 The payer hurdle in biotechnology's future.

Full size image

Dr Cliff Goodman, PhD, Vice President of the Lewin Group, believes that the payer bar is rising, and will be higher for products in early stage development and launched in 2012 or thereafter than it is now. Manufacturers will need to anticipate payers’ evidence requirements and incorporate them into their development plans.

Coverage and payment for biotechnology agents should not be ‘taken for granted’ and occur toward the end of a drug's development (Phase III). Instead, payer needs should be part of the clinical development process, and occur at several points in the process (Figure 3).

Figure 3
figure 3

 Incorporating payers into drug development.

Full size image

COMPOUND IDENTIFICATION

Payer needs and perspectives should be incorporated into the selection of clinical indications for development. For example, payers interviewed for this study identified the following disease states as on their radar screens, whether because of mortality, morbidity and/or direct costs to payers:

  • Pulmonary embolism/deep vein thrombosis (PE/DVT)

  • Hip fractures

  • STEMI myocardial infarctions

  • Myocardial infarctions (ST segment elevation [STEMI] and secondary prevention)

  • Stroke

  • Congestive heart failure (CHF)

  • Chronic obstructive pulmonary disorder (COPD)

  • Peripheral arterial disease (PAD)

  • Alzheimer's

  • Osteoarthritis (hip and other fractures)

  • Diabetes (macrovascular outcomes such as amputations and renal failure)

Even within these conditions, identifying events that lead to significantly higher morbidity, mortality or direct costs is important in compound development. For example, Medicare and Medicaid are major payers for stroke. A 2009 study4 examining 4 years of retrospective data found that Medicare spending in a cohort of patients with hemiparesis (weakness on one side) was $77 143 versus $53 319 in patients without hemiparesis (see end notes for more information about this study). If a biotechnology product substantially reduced hemiparesis, it could well have benefit in terms of direct costs, as well as morbidity and morbidity. At the same time, because of how the US reimbursement system is structured, if the agent was administered post the hospital admission (and thus received separate reimbursement), the manufacturer would be likely to be able to achieve higher payment. Alternatively, if the drug was administered during the in-patient hospital stay, if it provided substantial economic benefits such as reduction in costly complications or reduced length of stay, this would facilitate the argument for separate or additional payment.

Developing a very early understanding of payer needs and the reimbursement environment is particularly important for companies that are making ‘portfolio’ selection and allocation decisions. Pharmaceutical and biotechnology companies have many opportunities for research, acquisition or in-licensing, so understanding payer needs and the reimbursement environment is valuable from the perspectives of R&D focus, capital usage and ultimately revenues. Venture capital funding is limited, so identifying biotechnology agents that will have an attractive reimburse-ment profile is key to the venture capital firm's return on investment. Identifying how the biotechnology agent will provide value to payers and be successful in the reimbursement environment would also be valuable in raising additional funding or selling development and commercialization rights.

ASSESSING AND PLANNING FOR THE REIMBURSEMENT AND PAYER ENVIRONMENT

We recommend understanding the payer and the reimbursement environment be incorporated at several points in the clinical program. Dr Mike Murphy, MD, Chief Medical and Science Officer, Worldwide Clinical Trials, suggested that the process might begin at the time of the investigational new drug application or otherwise immediately after initial clinical data suggested good prospects for a clinical development program leading toward a specified indication. A key reason is to help design the clinical trial program before it is developed and shared with the FDA. Collection of payer-related data may be incorporated within the core clinical trials for regulatory approval, or in conjunction with the core trials, via a series of studies conducted sequentially or in parallel with the trials intended to secure regulatory approval.

From both a manufacturer and a regulatory perspective, it is much better not to launch the expensive and lengthy clinical development process unless the commercial landscape following approval can be anticipated and defined with data that will inform all stakeholders, and specifically not to make substantive changes to the clinical trial program during Phase III. Also, incorporating this evaluation at the end of Phase I or early in Phase II is more cost-effective for sponsors, and will help in designing Phase III trials and outcomes data collection correctly. John Avallanet, Managing Director and Principal of Cerulean Associates makes the point that early reimbursement evaluation (including volume of use and pricing) offers manufacturers ‘the chance to set better cost controls for production and the rest of development; it will hardly benefit you as the company if you are making a product that costs you more than you will earn back in reimbursement’.5

The payer environment and payer perceptions of data (or potential data) related to the product should also be assessed at the end of Phase II. One reason for this is that the payer environment is rapidly changing, and payers, payment mechanisms or payer priorities may change. Many other changes, such as the release of new comparative effectiveness research (CER) data (see later discussion), or development of new non-biotechnology interventions are also possible, and their potential impact should be assessed. Even if the new data are collected in an additional arm of the Phase III program, or outside the clinical program entirely, this provides the biotechnology agent's manufacturer with time to develop these data before or shortly after launch.

In Phase II and III trials, maintaining the highest possible standards for collecting payer related data (ideally as part of the random controlled trials for regulatory approval), or otherwise in parallel with trials for regulatory approval, is highly valued by payers.

Dr Saty-Murti,6 MD, noted that late stage changes in the clinical trial program could affect payers’ receptivity to covering the agent. For example if there is a change in Phase III of a clinical program in oncology from a primary (overall survival (OS)) to a secondary (progression-free survival (PFS)) endpoint based upon accumulating data, payers may well be sensitized to the prospect of a new (and expensive) intervention that may have less than optimal clinical benefit.

PAYERS VALUE CLINICALLY SIGNIFICANT OUTCOMES

Payers value outcomes data, and have a strong preference that these data come from prospective randomly controlled trials, partly for its objectivity, and also so that it can be evaluated independently. The data should be clinically meaningful and reproducible.

To understand what outcomes are meaningful from a payer perspective, it is essential to understand who the payers are, how the product (or disease state) affects outcomes that are important to payers, and payer cost or reimbursement amounts. Two examples of this are:

  • A pharmaceutical agent for overactive bladder reduced hip fractures which occurred when patients fell at night (assessment of the frequency and associated expenditures of hip fractures identified as an outcome of interest early in clinical development, and the clinical trial program included several years of patient follow-up) beyond that required technically for registration purposes.

  • A once yearly agent for osteoporosis that has an extremely high patient adherence rate, versus other, more frequently administered agents, which have much lower adherence (∼50 per cent).

Payers do not equate statistical significance with clinical importance and will critically evaluate the size of a purported effect and its impact across multiple dimensions of clinical care. The Medicare Evidence Development and Coverage Advisory Committee (MedCAC), for example, emphasizes that the improvement in outcomes should be clinically significant and generalizable to clinical practice. If a drug will have significant use in the Medicare population, it is important to include these patients in clinical trials.

Dr Aubry,7 MD, a national leader in health technology assessment and coverage decision-making in both the public and private sectors, commented that payers value data less when it measures a ‘proxy’ or an intermediate value, as opposed to a true outcome. For example, in cancer, tumor shrinkage and PFS are two measures that may not affect OS, the key primary outcome. Unless the surrogate, such as laboratory or radiologic results, can be strongly and directly linked to an important outcome in the peer-reviewed literature, proxy or intermediate values may be discounted by payers, particularly managed care.

The worlds of R&D and payers are truly different. Frequently, for clinical developers to understand what payer needs are and how to incorporate them into clinical trials, creative communication and ‘Translation’ from one ‘world’ to the other is needed; that is, a partnership of those with key questions with individuals capable of accessing tools and creating data potentially providing answers. Dr Murphy commented that a key issue within the development sphere is to understand ‘what to measure’ – what outcomes are most relevant to payers. Deciding how to incorporate assessment of those outcomes into the overall program design becomes critical to designing clinical trials that will support payer reimbursement.

THE INCREASING IMPORTANCE OF COMPARATIVE DATA

Dr Aubry noted the growing importance of comparative data, and of data that show significant improvement in outcomes. Dr Goodman observed that, in the future, for products that demonstrate clinical effectiveness or cost effectiveness, there may in fact be greater opportunity. He noted a ‘payoff in getting over the hurdles’.

In interviews with managed care decision-makers, they have said that in established therapeutic classes, including biologics such as the anti-TNFs, they have seen very little clinical differentiation, and the differences have not been significant from a payer decision-making perspective. Even now, the typical payer evaluation process for new biotechnology agents in an existing class may be truncated. In the future, there may well be less opportunity for ‘me too’ products, unless they have a definite niche, or the manufacturers are willing to be flexible in their pricing or contracting.

In contrast, when managed care payers perceive that a biotechnology agent is substantially different from other agents in an existing class, or when it is the first agent in a new class, the importance of clinical data, particularly efficacy, becomes much more important (Figure 4).

Figure 4
figure 4

 Managed care decision-making for differentiated agents.

Full size image

When new biotechnology agents enter a class, the managed care committees that decide the new agents’ coverage and reimbursement highly value head-to-head clinical trials, so that the comparison is objective and direct, and how the trial was conducted and the comparative data developed, can be fully understood. Another benefit of having these data at launch, rather than in a post-marketing trial, is that payer decision-makers can then use the data in decision-making during their first review, which is typically more in-depth and when perceptions are formed, and it can benefit the drug in early commercialization. Managed care decision-makers report that they are seeing more head-to-head data in clinical trials; including head-to-head data will be even more important in the future.

From a payer perspective, another issue may occur if a new biotechnology agent is to be used in addition to existing biotechnology agents. A Medical Director from a national payer commented that typically the incremental value of second or third line biotechnology agents is less than the first agent, but the total cost is of course substantially higher. Conversely, several managed care decision-makers commented favorably if combination therapy is replaced by monotherapy.

THE POTENTIAL IMPACT OF CER

Another potential consideration for biotechnology manufacturers relates to the US$1.1 billion for CER included in the American Recovery & Reinvestment Act of 2009. The CER program's focus is between different types of interventions; it is not between agents in a given therapeutic class (for example, between biologics for rheumatoid arthritis). Some examples of the first top 100 CER priorities8 selected by the Institute of Medicine include:

  • compare the effectiveness of treatment strategies for atrial fibrillation including surgery, catheter ablation and pharmacologic treatment;

  • compare the effectiveness of treatment strategies for vascular claudication (for example, medical optimization, smoking cessation, exercise, catheter-based treatment, open surgical bypass);

  • compare the effectiveness of treatment strategies (for example, artificial cervical discs, spinal fusion, pharmacologic treatment with physical therapy) for cervical disc and neck pain.

Based on CER's focus on comparing different interventions, potentially promising areas for biotechnology target selection is versus other types of interventions that are ineffective, have high morbidity or mortality or are very expensive.

Notably, the top 100 CER priorities do include several biotechnology agents, so the biotechnology industry should be alert to this possibility, particularly if there are not strong outcomes data for the use of biologics to treat that condition.

  • compare the effectiveness of different strategies of introducing biologics into treatment algorithms for inflammatory diseases, including Crohn's disease, ulcerative colitis, rheumatoid arthritis and psoriatic arthritis.

  • compare the effectiveness (including quality of life) of treatment strategies (for example, topical steroids, ultraviolet light, methotrexate, biologic response modifiers) for psoriasis.

  • compare the effectiveness of different treatment options (for example, laser therapy, intravitreal steroids, anti-vascular endothelial growth factor (anti-VEGF) for diabetic retinopathy, macular degeneration and retinal vein occlusion).

Federally funded CER is not allowed to include cost in its calculations, but once data for a particular CER priority are available, other groups, such as technology assessment groups or health-care policy organizations, or even individual health economists, can be expected to integrate information about costs into the CER results, and publicize these data. Of course, managed care organizations (particularly larger plans, and more sophisticated plans) may also integrate CER results and actual plan costs and use these data in their coverage and reimbursement decision-making. Pharmacy Benefit Managers may also integrate these data and share their findings and recommendations with their managed care, employee benefit management, employer clients.

CONCLUSION

Orienting R&D programs to payer priorities from an early point is likely to lead to development of agents that have substantially higher volume of use, driven by improvements in payer coverage and reimbursement, patient access and physician prescribing. Notably, the cost of incorporating payers into the R&D process is substantially less than the potential revenue gain. Additionally, incorporating the development of objective comparative data (versus existing biotechnology agents for that indication or other non-biologic interventions for the disease state) in the pre-launch clinical trial program will also be increasingly important, as payers are likely to use these data in their coverage and reimbursement decision-making.