COA strongly opposes any proposal to incorporate middlemen into the Part B program and empowering those entities to wield influence on cancer treatment. We firmly believe that all Medicare seniors, especially those facing life-threatening diagnoses like cancer, should have timely and unhindered access to the therapies that their physicians believe to be most effective for them. We do not believe that allowing corporate middlemen – who do not have in-depth clinical oncology experience – into Part B under CAP will improve patient care or outcomes. In fact, experience documents that it will do the exact opposite.
Under the IPI Model, physicians would be limited in their ability to treat patients to the best of their medical knowledge and experience, as CAP vendors would be able to directly or indirectly dictate the choice of therapies in the model. As cancer treatment becomes more precise and individualized, introducing an additional layer between patients and life- saving medicines is not only inappropriate, it could be very dangerous for seniors facing life-threatening diagnoses.
Furthermore, we are concerned that adding middlemen to Part B through the IPI Model would introduce the same cancer treatment access challenges experienced by cancer patients today in Part D. COA is befuddled by the administration’s interest in increasing the power and prevalence of middlemen at a time when both policymakers and the public have a heightened awareness of the business model and distorted incentives that have allowed PBMs to consolidate and grow their influence without reducing costs. In Medicare Part D, PBMs have increasingly sought to control costs by using a variety of mechanisms to steer patients toward less expensive therapies, but ultimately unduly complicating drug procurement, delaying and denying patients’ treatment, and driving up costs.
COA believes that the IPI Model could repeat past reform mistakes. The original CAP program, which ran for 18 months between 2006 and 2008, has been suspended for the last decade for good reasons. The original program presented a host of issues for Medicare, including negative impacts on access and supply chain dynamics and limited participation from both physicians and vendors. The original CAP program, we note, did not effectively lower costs for Medicare or for patients, and added significant complexities to drug negotiations and procurements. If CMS decides to reimpose CAP on the Medicare program, it will force the creation of individual patient inventories, increase the likelihood of treatment errors and delays, and place new and unreimbursed administrative burdens on clinicians. While the agency technically has the statutory authority to bring back CAP, COA strongly urges the administration to reconsider this proposal. CAP was not the answer to increasing competition in Part B in 2008, and it is not the answer in 2018.
Concerns About Included Drugs and Biologicals in the IPI Model as Proposed
It is unclear which drugs or therapeutic classes would ultimately be included under the proposed demonstration, but CMS has indicated its interest in including the “top spending” Part B drugs in the IPI Model. The agency has stated that it would likely start with an initial set of drugs in the early model years before expanding to a broader set of therapies.
COA believes that the IPI Model proposal should be withdrawn, but at a minimum cancer drugs should be excluded. Cancer treatment is a complex and sensitive therapeutic area, and it is inappropriate to subject patients with cancer and the providers that treat them to large-scale experimentation that could threaten patient outcomes.
Since the Secretary has the authority to exclude any drugs and biologicals whose inclusion is unlikely to result in cost savings or whose inclusion would have an adverse effect on access, we would like to flag some issues related to access to cancer drugs under the original CAP7 to point to the need to carve oncology out of any new CAP demonstration, regardless of how it is modified:
- On average, among the Top 30 CAP drugs in 2007, CAP payment for cancer drugs was higher than ASP plus 6%.
- Between 20-30% of CAP claim line items for cancer drugs were billed under the “emergency restocking” provision to account for point-of-care changes to previously ordered drugs based on patients’ clinical presentation at the time of the visit. This means that CAP-participating oncology practices had to maintain a stock of drugs for emergency use at some financial risk for about 1 of every 4 drugs they administered.
- Provider surveys on overall satisfaction with CAP among participants showed that oncology specialists reported being less satisfied than other providers surveyed, likely due to the complexity of the therapeutic area, the types of drugs that they administered, and the types of Medicare beneficiaries who they treated.
All Medicare seniors should have access to all therapies available, and the IPI Model as proposed would severely restrict the use of both new and standard-of-care cancer treatments. Cancer patients rely heavily on innovation in treatments and COA is concerned that every element of this model seriously challenges access to therapies and innovation. As the newest and innovative therapeutic products are, by definition, single-source, this proposal could limit the availability of the best available options in cancer treatment for seniors and other Medicare beneficiaries in the Part B program. Furthermore, CMS’ proposal could disincentivize lifesaving innovations.
COA believes the current Part B system offers the flexibility to use and administer the most clinically appropriate therapies based on clinical considerations, without delays or disruptions in cancer care.
Concerns About Flawed Notions of Part B Prescribing in the IPI Model as Proposed
The focus on changing the provider payment for drugs is grounded in the flawed notion that providers choose treatments based on financial considerations, rather than clinical ones. This is both incorrect and offensive. Independent, peer-reviewed research has found that modifying payment for cancer care does not impact clinical decision making for providers, but rather that the factors that actually impact prescribing patterns are the introduction of new drugs, new clinical evidence, and identifying new best-practices for treatment.8 In short, medical professionals make medical decisions using clinical evidence, not for financial gain. Providers already face the burdens associated with identifying and educating themselves on innovative techniques and therapies that can help their patients. Employing bureaucratic barriers to accessing these therapies is inappropriate.
Attached is a just completed meta-analysis on the subject of oncologist prescribing (Attachment B: “The Myth of Perverse Physician Incentives: Examining Research and Accusations in the Medicare Part B ASP Reimbursement System for Oncology”). It thoroughly debunks the false narrative that oncologists’ decision-making is driven by financial factors, much of which is based on extremely outdated and/or fundamentally flawed research. In fact, a number of studies on physician prescribing under the Part B reimbursement system have been published in recent years, all of which reach dramatically differing conclusions on physician prescribing patterns. They identify the following as drivers of drug selection by oncologists:
- Highest-quality patient care;
- Most effective treatment options;
- Best expected outcome;
- Best tolerated, least toxicity or permanent ill effects;
- Care regimen least disruptive to daily life;
- Most cost-effective option for the patient;
- Financial cost to the patient; and
- Changing physician and patient expectations about cancer care.
Concerns About the IPI Model as Proposed and Interactions with Other Models and Federal Programs
CMS requested comments on how to avoid unintended consequences of the interaction between the IPI Model as proposed and other federal programs, as well as the potential impact of the inclusion of manufacturer sales to model vendors on “best price” and average manufacturer price (“AMP”), and whether CMS should exempt prices offered under the model from AMP and “best price” calculations.
It is important that any Part B measures not disrupt or further complicate existing value-based care initiatives. While COA is totally committed to increasing the quality of cancer care and decreasing costs, as well as supporting value-based models that can positively impact prices and utilization, the IPI Model as proposed is not value-based. The IPI Model appears to be designed to target only costs and gives no consideration to how it will impact quality of care, access to treatment, or the utilization of those treatments.
COA is already working to spearhead innovative payment reforms aimed at lowering costs and improving the quality of cancer care, notably including the OCM. Both COA and CMS have invested heavily in the OCM and we urge CMS to consider the impact that the IPI Model as proposed would have on patients that are already benefitting from the OCM and other ongoing commercial value-based payment models. We believe that the threats to the OCM presented by the IPI Model could represent an unfortunate loss of years of effort, resources, and funds if the IPI Model as proposed is implemented.
Concerns with the Indexing of Part B Drug Reimbursement to International Prices
To reiterate, we are very concerned about the increasing cost of Part B drugs based on increasing prices. However, we do not believe that the international indexing component of the IPI Model is a realistic solution to decreasing Part B drug costs. We are concerned about the feasibility of indexing drugs to international prices, about the impact on all cancer drug reimbursement (Medicare fee-for-service, MA, and commercial), and about the operational implementation of index pricing.
In terms of the feasibility of indexing cancer drug prices to international prices, we are concerned about the consistent availability of cancer drugs across all countries to arrive at consistent pricing. In a number of the countries that would be involved in the indexing basket of drugs, the same drugs available in the U.S. are not available in those countries – or are available for a limited set of indications or are available under different names. Additionally, manufacturers may possibly increase international “list” prices while using rebates and discounts to bring down the true “net” price of a drug in other countries. As it is, we believe that U.S. “list” drug prices are forced higher by the increasing scope and magnitude of required rebates and discounts, the latter notably including discounts required under the 340B Drug Pricing Program (“340B”). The increasing scope and magnitude of 340B drug discounts are fueling “list” prices of Part B drugs. Equating U.S. “net” drug prices (i.e., ASP) with other countries, where the “list” price may not be reflective of rebates and discounts, creates “apples and oranges” pricing.
The problems just discussed may create a situation wherein the arrived at IPI Model index price is artificially and unrealistically low. Although at first blush this appears to benefit Medicare and its beneficiaries by lowering prices, this is a house of cards that is not realistic and sustainable. There is no way a CAP middleman vendor, or even a provider, will be able to use an artificially low international price to leverage that price, or lower, from a manufacturer, given that most oncology drugs do not have competitive substitutes (either generic, biosimilar, or brand therapeutic substitutes). What happens then? Will CMS reimburse the CAP vendor (or the provider, if not CAP vendor) for a drug at the index price, even though the vendor, or the provider, cannot purchase the drug anywhere close to that price? No CAP “vendor” will participate in the program under that scenario and forcing a provider to accept drug reimbursement for less than what a drug can be purchased for is simply not financially feasible.
Additionally, we note that any models involving ASP-based reimbursement must be exempted from manufacturers’ calculation of ASP so as to not inappropriately bring down the ASP calculation for those not participating in the proposed IPI Model, or other models. And CMS must not apply the 2% Medicare sequestration payment cut to any drug reimbursement models that are tested.
Finally, we note that “price fixing” of any goods and services has never worked in the long-term in the history of the world and typically leads to shortages, which already is a problem with low-priced generics. There are more effective and sustainable ways of lowering drug prices/costs through competition and market reform.
Comments on Alternatives to the IPI Model Components
As we have discussed, community oncologists are very concerned about access to and affordability of treatment for our patients, as well as the viability of our practices to continue providing the highest quality, most affordable, and most accessible cancer care. These concerns have guided us in analyzing the proposals in the IPI Model and in developing and analyzing possible alternatives. As we referenced earlier in this letter, COA is working on possible alternatives including the following:
- Tiering ASP-Based Reimbursement: COA highly supports the drug reimbursement formula established in the MMA under careful deliberations between Congress and stakeholders. We believe that an add-on percentage versus a flat fee (for example, based on some historical ASP, as proposed in the IPI Model) is important to preserve for two reasons. First, as we will explain, we are analyzing a modification to the percentage-based fee that is tiered as opposed to the current add-on fee of plus 6% (which, it should be noted, is really 4.3% after application of the 2% Medicare sequester cut) that we believe will be more appropriate to account for escalating drug prices and the emerging biosimilar market. Second, any type of flat fee will never keep up with medical inflation and the costs relating to the increasing complexity of Part B drug procurement, storage, and handing.
We note, as COA has done on numerous occasions, that the add-on to ASP is not “profit,” as some falsely assert. It has to cover the very real increasing costs of human resources and infrastructure required to procure, handle, store, inventory, and dispose of chemotherapy and other cancer-related Part B drugs. What is left over covers operating expenses and bad debt. As it is, in the Medicare Physician Fee Schedule for 2019, community oncology practices will be reimbursed 3% less for administering chemotherapy. This is nonsensical and illustrates the problems of flat fees determined by Medicare formulas updated annually.
What COA is considering and analyzing, and recommends that CMS considers, is an alternative to a flat fee – a tiered percentage on ASP. More specifically, we are analyzing the following tiers of ASP add-on payments:
- A very high percentage add-on for generics;
- A higher than 6% add-on for biosimilars; and
- A tiered structure for brands whereby higher priced brands have a lower add-on payment and the add-on payment increases as the brand price is lower.
Note that we are not suggesting in any way that finances dictate oncologists’ choices in making treatment decisions in consultation with their patients. However, given both the escalating cost of brand drugs and what will be the introduction of more biosimilars in oncology, a tiered add-on will create a more appropriate reimbursement structure for what the add-on to ASP covers. Additionally, with biosimilars the goal is to create a robust and healthy market. With generics, increasing reimbursement will substantially help mitigate the constant shortages plaguing the market, as FDA Commissioner Scott Gottlieb, MD recently suggested.9
We also note that a tiered add-on to ASP, coupled with the clinically appropriate utilization management concept outlined below, will impact manufacturers’ pricing decisions for newly launched therapies depending on their degree of innovativeness. COA wants to maintain and reward drug therapy innovation and to continue to have access to new innovative therapies for our patients, but while controlling the costs of new therapies.
Our goals in considering this tiered ASP percentage add-on, in conjunction with employing clinically appropriate utilization management (as described below),
- Maintain access to innovative new cancer therapies and incentivize manufacturers to continue to introduce those therapies;
- Enhance oncologists’ clinically appropriate decision making;
- Control the pricing and costs of cancer drug therapies for patients and Medicare (and the taxpayers who fund the program);
- Create a robust and healthy biosimilar market;and
- Maintain a more healthy, competitive generic market not plagued by constant shortages.
We realize that more detail is needed in this proposal, but the timing of this comment letter submission deadline only allows us to detail this proposal as much as we can at this point.
- Employing Clinically Appropriate Utilization Management (“CAUM”) in Part B: If the administration is determined to find ways to steer prescribing toward the highest value therapies in Part B, it should do so by giving the right tools and incentives to physicians themselves, instead of PBMs and other middlemen CAP vendors. A potential CAUM program that incentivizes the use of pathways and clinical protocols developed by physicians would protect evidence-based care much more effectively than other entities whose decisions are purely based on cost. COA has been engaging with other stakeholders and we see a great opportunity to empower providers to ensure the most appropriate and value-based prescribing, without jeopardizing patient access or timely care. Furthermore, this would also help community oncology practices to more easily move toward sophisticated, performance-based alternative payment models.
We believe that a CAUM provider-led model would be patient-centric, but at the same time, it would provide CMS an opportunity for savings by ensuring that Part B drug utilization best reflects value and efficacy, while encouraging more competition between manufacturers. Coupled with a tiered ASP add-on percentage, as outlined above, CAUM would create an environment that drives both manufacturer research and development, as well as oncologist decision making, in collaboration with patients, towards high value (a function of both quality and cost). Manufacturers would be driven by value in research and development, both in terms of therapy innovation and pricing. Oncologists and other Part B providers would be driven by the same value in adhering to provider-developed pathways, with Part B drug reimbursement equitable for generics, biosimilars, and brands. Finally, this would be in an environment devoid of any PBMs and related middlemen dictating treatment decisions based on rebates and other financial incentives feeding their bottom line profits rather than on what is clinically appropriate and in patients’ best interests.
- Addressing High Out-of-Pocket (“OOP”) Costs: As providers, we are well aware that cancer patients and survivors face serious hardships and anxiety related to the affordability of their cancer treatments. COA is very supportive of efforts to reduce OOP costs for Medicare beneficiaries by reducing the coinsurance amount, introducing an out-of-pocket maximum, or shifting to fixed copays for drugs. Part B fee-for-service beneficiaries pay a 20% coinsurance on all medical services, including drugs. Even though most patients have some type of supplemental medical insurance (e.g., Medigap, employer plans), those Part B beneficiaries without supplementary coverage face the greatest exposure to drug costs under Medicare. Seniors with pre- existing conditions, such as those who get diagnosed with cancer, can be denied Medigap coverage in many cases when they apply for supplemental insurance policies, leaving them exposed to the full burden of coinsurance on very costly therapies. We understand that in this fiscal environment this policy could increase Medicare costs, but we believe that there are multi-stakeholder solutions with support from manufacturers that can produce rebate passthrough or additional copay assistance mechanisms for low-income beneficiaries or for those who were denied supplemental insurance due to health status.
- Lowering Drug Prices Without Artificial International Price Indexing: We have serious concerns about the indexing of Part B drug reimbursement to international prices, as explained earlier in this letter. For close to a year (well before the IPI Model was proposed), COA has been in discussions with well over 20 manufacturers on the concept and execution of performance-based pricing arrangements (e.g., indication and outcomes pricing models) as part of our development of the OCM 2.0. Although these types of arrangements (which CMS has referenced in relation to CAP vendors in the IPI Model) show real promise in addressing payment for drugs based on value, they would require waivers from Medicaid best pricing, ASP, etc. to be implemented in the Medicare system. We welcome the opportunity to discuss this further with CMS. Additionally, since the release of the IPI Model proposal, COA has been in discussions with several manufacturers on alternatives to international indexing, which involve rebating and other creative approaches. We are not at liberty to disclose the specifics of those alternatives at this time in this letter, but we are hoping you will hear directly from manufacturers.
Conclusion
As we have detailed very extensively in this letter and the supporting attachment, COA cannot support any mandatory experiment on patient care using CMMI as the IPI Model proposes. Rather, we hope to be able to continue working with CMS on constructing a patient-centric, value-driven, voluntary demonstration project with appropriate patient safeguards. COA would enthusiastically support this and help drive large-scale participation by community oncology practices across the country. The OCM provides the perfect model of collaboration whereby CMMI and providers have been working together to develop, implement, and refine a voluntary model of oncology payment reform.
Although COA has serious concerns with the IPI Model as proposed, we are very actively developing and analyzing alternative approaches. This should be demonstrated by our meetings and calls with CMS leadership and staff, as well as the comments provided in this letter. If you examine COA’s work over the past eight years, and the specific programs implemented by practices, community oncology has done more to advance payment reform in general and very specifically in cancer care. This dedication to doing what is right has come at a time when community oncology practices have faced declining Medicare reimbursement; enormous obstacles to providing clinically appropriate care, such as increasing inappropriate prior authorizations by insurers and other middlemen; and the destructive and harmful behavior of PBMs in obstructing cancer patients from getting timely and appropriate cancer care.
COA’s unwavering commitment and steadfast determination to continually improve our cancer care system is driven by a mission to ensure that patients with cancer, the majority of which are seniors who are Medicare beneficiaries, continue to have access to the highest quality, most affordable, and most accessible cancer care in the communities where they live and work.
COA looks forward to working closely with CMS to advance meaningful, patient-centered, and value-driven policies relating to cancer care. We are available to discuss any of our concerns and recommendations provided in this letter.
We thank you for your consideration.
Sincerely,
Jeff Vacirca, MD
President
Michael Diaz, MD
President Elect
Ted Okon
Executive Director
COA Comments on IPI for Medicare Part B Drugs
The Honorable Seema Verma, Administrator
Centers for Medicare & Medicaid Services
Department of Health and Human Services
Attention: CMS-5528-ANPRM
P.O. Box 8016
Baltimore, MD 21244-8013
Re: Medicare Program; International Pricing Index Model for Medicare Part B Drugs; CMS-5528- ANPRM
Dear Administrator Verma:
On behalf of the Board of Directors of the Community Oncology Alliance (COA), we are submitting this comment letter regarding the Medicare Program; International Pricing Index Model for Medicare Part B Drugs (CMS-5528-ANPRM) (the “IPI Model”).
As you know, COA is an organization that is dedicated to advocating for the complex care and access needs of cancer patients and the community oncology practices that serve them. COA is the only non-profit organization in the United States dedicated solely to independent community oncology practices, which serve the majority of Americans receiving treatment for cancer. COA’s mission is to ensure that cancer patients receive quality, affordable, and accessible cancer care in their own communities where they live and work. For more than 15 years, COA has built a national grassroots network of community oncology practices to advocate for public policies to support cancer patients.
First, we want to say that we appreciate your openness and interest in understanding our position on this important topic of the cost of prescription drugs in the Medicare Part B program (“Part B”), as well as the overall total cost of cancer care, which is comprised of drugs and services, both hospital and physician related. We thank you for meeting with us on two occasions leading up to the preparation and submission of this comment letter, as well as the opportunity to submit this letter. Although we are very concerned about escalating drug prices/costs and the high total cost of cancer care, as we have expressed to you, COA does not support the IPI Model as proposed in the pre- proposed rule published by the Centers for Medicare & Medicaid Services (“CMS”) because we have serious concerns about its impact on cancer patient care and even its legality. That said, we are actively working on solutions to high drug costs and the total cost of cancer care and, as we outline in this comment letter, are developing and analyzing alternatives to the model.
COA is sincerely committed to oncology payment reform that makes cancer care more affordable, as witnessed by our commitment and involvement in the Oncology Care Model (“OCM”); our ongoing development of the OCM 2.0, a next-generation, more universal oncology payment model; the numerous summits and meetings we have held with payers, employers, stakeholders, and providers; and the involvement of community practices, like ours, in an incredible number of private insurance payment models and programs. Community oncologists are concerned about the escalating prices and costs of cancer drugs, as well as high hospital-related costs, all of which contribute to the overall increasing total cost of cancer care. As leaders in the delivery of cancer care, we are mindful of our responsibility to be good stewards of costs we can control, including the utilization of drugs and services.
In this letter, we will detail our problems and concerns regarding the IPI Model as proposed, even though we have summarized these with you, leadership, and staff from CMS and the CMS Innovation Center (“CMMI”) in previous exchanges. We understand and appreciate that CMS appears to be already considering changes to its proposed model, and implementation of it, based on input from COA and providers. However, we believe it is important to detail our concerns with the IPI Model as proposed, for both the sake of memorializing our comments and as understanding the basis for our alternative suggestions.
In this comment letter, we will also outline alternatives to the IPI Model, which we are currently working on in greater detail. Some of that detail will not be available in this comment letter, given the submission deadline of 12/31 and the holiday season, coupled with the fact that we have much more analytical work to do on these alternatives.
We hope to continue the dialogue with CMS post submission of this comment letter, as we continue our analytical process and CMS moves towards a proposed rule on the IPI Model.
Summary Comments
COA does not support the IPI Model, and we strongly urge CMS not to move forward with it as proposed. We are greatly concerned that the IPI Model, which would conduct nothing short of a mandatory national experiment on Medicare Part B beneficiaries, could disrupt access to the innovative therapies and care that vulnerable seniors with cancer and other serious diseases need and are guaranteed under Medicare. The reasons for COA’s position are summarized as follows:
We also note, for the record, that COA is adamantly against any mandatory demonstration projects conducted by CMMI. As we detail in the attached legal analysis (Attachment A: Legal and Constitutional Problems with a Mandatory International Pricing Index Model), mandatory demonstration projects are clearly not in the charter of CMMI as written into law by Congress. If CMS believes that CMMI has the power via statute to effectively amend Medicare law – in this case the Part B drug reimbursement rate for at least 50% of Part B providers – it (CMS representing the Executive branch) either has overstepped its constitutional boundaries separating the powers of government branches or Congress has effectively handed over its powers to the Executive branch. That would either be illegal or unconstitutional, with the latter case invalidating the section of the law that created and funded CMMI.
By their nature, “mandatory” demonstrations are such because they force providers to participate and do not have stakeholder input or buy-in. This was the case with the attempt by the previous administration to force 75% of Part B providers to participate in the failed Medicare Part B Purchasing Model in late 2016.
Finally, we note two other concerns with the IPI Model as proposed. First, we have concerns about operationalizing an international pricing index as Part B drug reimbursement and about the adverse impact on commercial reimbursement, as further detailed in the aforementioned attached document (Attachment A). Second, we have several concerns about a “flat” average sales price (“ASP”) add-on, or any type of “capitated” fee payment, which we will also touch on later in this comment letter.
In summary, COA is very concerned that the IPI Model as proposed will totally upend how cancer care is delivered under Part B given the proposed mandatory national scope. The introduction of middlemen to Part B will seriously threaten the access of seniors and other Medicare beneficiaries to cancer treatment currently being delivered on an immediate “just in time” basis and close to home. Additionally, we are concerned with how such widespread changes (impacting 50% of Part B providers) to Part B reimbursement will impact the viability of community cancer clinics to continue providing care.
COA’s concerns about the IPI Model as proposed are based not on supposition or unfounded threats but on two clear facts. First, as we have documented in four volumes of actual patient cases1, PBMs and related middlemen are having an increasingly adverse effect on the treatment of cancer patients under Medicare Part D and commercial pharmacy benefit plans providing oral cancer drugs. Second, as we have documented along with others, there has been a marked shift of cancer care from physician-run community sites into large hospital systems due to declining reimbursement, especially in Part B, and clinic closings in rural areas. Data on this shift can be clearly seen in the annual COA Practice Impact Report, which tracks the changing landscape of community cancer care in the U.S. In the 2018 report, we note that 1,653 community oncology clinics and/or practices have closed, been acquired by hospitals, undergone corporate mergers, or reported that they are struggling financially. An average of 3.5 community oncology practices have closed per month, a rate that remains unchanged since the last report issued in 2016. Overall, 13.8 practices per month have closed, been acquired by hospitals, or undergone mergers since 2008.2 A recent feature in the Houston Chronicle newspaper makes this trend of closures real, telling the story of the anticipated closing of the Rio Bravo Cancer and Blood Clinic, which is currently providing cancer care to patients from over a 10,000 square mile area in rural Texas. As related in the story, “Most patients are seniors and dependent on Medicare…. But the clinic has been in the red for about 18 months because of a series of Medicare cuts for cancer drugs….”3
Radically changing Medicare reimbursement on a national basis, without studying its impact, could accelerate the shift in the site of cancer care from independent community cancer clinics to much more expensive hospital systems, resulting in higher costs for Medicare, commercial insurers, and beneficiaries, as well as in serious cancer care access issues, especially in rural or underserved areas.
As we have stated, COA appreciates that CMS is actively soliciting stakeholder input on what is a proposed model in a pre-proposed rule. Also, we are encouraged that CMS is listening to alternatives and other approaches.
In terms of specific alternatives that we are considering and analyzing, they include:
Comments on Concerns with the IPI Model as Proposed
Opposition to the Mandatory National Scope of the IPI Model as Proposed
The IPI Model as proposed is one of the largest and most complex CMMI demonstrations developed to date, with a broad geographic reach and strong potential to disrupt both the drug supply chain and the care delivery system. COA is staunchly opposed to mandatory models, especially of the scale and complexity outlined as part of the IPI Model. As previously expressed, we are against using mandatory demonstrations as a vehicle for the Executive branch to effectively bypass Congress by making sweeping changes to Medicare law. The IPI Model establishes a “demonstration” of the size and scope that far exceeds anything that can be reasonably considered a “test.”
COA’s legal and constitutional reasons for our opposition to mandatory CMMI demonstration projects are summarized as follows:
Attached (Attachment A) is a detailed overview of the legal and constitutional problems with the IPI Model as proposed.
COA is also very concerned that the mandatory nature of the IPI Model and the completely randomized process to divide the country into experimental and control groups would create insurmountable operational changes and would disadvantage patients based on geography. We do not believe that the agency can arbitrarily select geographies without creating significant disparities in access to care for seniors and other Medicare beneficiaries with cancer. Further, because the entire supply chain for oncology products would be disrupted, the IPI Model will impact providers both inside and outside the demonstration. As such, the IPI Model will impact not only 50% of Part B, but also Medicare Advantage (“MA”) and commercial plans and beneficiaries – in effect, the entire country. Additionally, this will present serious treatment dilemmas for practices, providers, and patients, and does not represent a solid experimental design. Any wholesale changes to Part B, especially impacting care delivery, should first be carefully tested on a limited scale, through programs that are voluntary for both patients and providers, in order to avoid creating operational complexities, increasing administrative burdens, and potentially even adding more costs to Medicare and its beneficiaries.
We believe that the IPI Model would hurt independent community oncology providers, who, ironically, do not represent significant cost burdens to the Part B program. According to the Medicare Payment Advisory Commission (“MedPAC”), Part B spending has been growing more rapidly for hospital outpatient departments (“HOPDs”) than for physicians and suppliers. Between 2009 and 2015, Part B drug spending grew at an average of roughly 16% for HOPDs and about 7% for physicians.5 Similarly, a recent peer-reviewed study found that both the mean total cost of care and the mean chemotherapy costs per patient per month were significantly lower for patients treated in a community-based practice compared with those treated in a hospital-based outpatient facility.6
We understand that the administration sees the increase in spending on oncology therapies as a challenge for patients and the federal budget. However, we do not agree that it is appropriate to address Part B cost growth by targeting community practices and independent physicians and endangering their ability to practice, especially given their proven role in driving value and efficiency in cancer care. Furthermore, cancer is a complex and sensitive therapeutic area and it does not lend itself well to models that seek to interfere with evidence-based clinical decision-making. Oncologists should not be punished for embracing medical innovation in the form of immunotherapies, targeted therapies, and personalized medicines that have transformed care and given patients with cancer a fighting chance.
Concerns about CAP and Middlemen as CAP Vendors in the IPI Model as Proposed
Under the IPI Model as proposed, CMS is looking to leverage its dormant authority under the previous failed CAP to contract with private-sector “vendors” to act as middlemen between manufacturers, physicians, HOPDs, and other providers for the drugs and biologic products included in the IPI Model.
COA strongly opposes any proposal to incorporate middlemen into the Part B program and empowering those entities to wield influence on cancer treatment. We firmly believe that all Medicare seniors, especially those facing life-threatening diagnoses like cancer, should have timely and unhindered access to the therapies that their physicians believe to be most effective for them. We do not believe that allowing corporate middlemen – who do not have in-depth clinical oncology experience – into Part B under CAP will improve patient care or outcomes. In fact, experience documents that it will do the exact opposite.
Under the IPI Model, physicians would be limited in their ability to treat patients to the best of their medical knowledge and experience, as CAP vendors would be able to directly or indirectly dictate the choice of therapies in the model. As cancer treatment becomes more precise and individualized, introducing an additional layer between patients and life- saving medicines is not only inappropriate, it could be very dangerous for seniors facing life-threatening diagnoses.
Furthermore, we are concerned that adding middlemen to Part B through the IPI Model would introduce the same cancer treatment access challenges experienced by cancer patients today in Part D. COA is befuddled by the administration’s interest in increasing the power and prevalence of middlemen at a time when both policymakers and the public have a heightened awareness of the business model and distorted incentives that have allowed PBMs to consolidate and grow their influence without reducing costs. In Medicare Part D, PBMs have increasingly sought to control costs by using a variety of mechanisms to steer patients toward less expensive therapies, but ultimately unduly complicating drug procurement, delaying and denying patients’ treatment, and driving up costs.
COA believes that the IPI Model could repeat past reform mistakes. The original CAP program, which ran for 18 months between 2006 and 2008, has been suspended for the last decade for good reasons. The original program presented a host of issues for Medicare, including negative impacts on access and supply chain dynamics and limited participation from both physicians and vendors. The original CAP program, we note, did not effectively lower costs for Medicare or for patients, and added significant complexities to drug negotiations and procurements. If CMS decides to reimpose CAP on the Medicare program, it will force the creation of individual patient inventories, increase the likelihood of treatment errors and delays, and place new and unreimbursed administrative burdens on clinicians. While the agency technically has the statutory authority to bring back CAP, COA strongly urges the administration to reconsider this proposal. CAP was not the answer to increasing competition in Part B in 2008, and it is not the answer in 2018.
Concerns About Included Drugs and Biologicals in the IPI Model as Proposed
It is unclear which drugs or therapeutic classes would ultimately be included under the proposed demonstration, but CMS has indicated its interest in including the “top spending” Part B drugs in the IPI Model. The agency has stated that it would likely start with an initial set of drugs in the early model years before expanding to a broader set of therapies.
COA believes that the IPI Model proposal should be withdrawn, but at a minimum cancer drugs should be excluded. Cancer treatment is a complex and sensitive therapeutic area, and it is inappropriate to subject patients with cancer and the providers that treat them to large-scale experimentation that could threaten patient outcomes.
Since the Secretary has the authority to exclude any drugs and biologicals whose inclusion is unlikely to result in cost savings or whose inclusion would have an adverse effect on access, we would like to flag some issues related to access to cancer drugs under the original CAP7 to point to the need to carve oncology out of any new CAP demonstration, regardless of how it is modified:
All Medicare seniors should have access to all therapies available, and the IPI Model as proposed would severely restrict the use of both new and standard-of-care cancer treatments. Cancer patients rely heavily on innovation in treatments and COA is concerned that every element of this model seriously challenges access to therapies and innovation. As the newest and innovative therapeutic products are, by definition, single-source, this proposal could limit the availability of the best available options in cancer treatment for seniors and other Medicare beneficiaries in the Part B program. Furthermore, CMS’ proposal could disincentivize lifesaving innovations.
COA believes the current Part B system offers the flexibility to use and administer the most clinically appropriate therapies based on clinical considerations, without delays or disruptions in cancer care.
Concerns About Flawed Notions of Part B Prescribing in the IPI Model as Proposed
The focus on changing the provider payment for drugs is grounded in the flawed notion that providers choose treatments based on financial considerations, rather than clinical ones. This is both incorrect and offensive. Independent, peer-reviewed research has found that modifying payment for cancer care does not impact clinical decision making for providers, but rather that the factors that actually impact prescribing patterns are the introduction of new drugs, new clinical evidence, and identifying new best-practices for treatment.8 In short, medical professionals make medical decisions using clinical evidence, not for financial gain. Providers already face the burdens associated with identifying and educating themselves on innovative techniques and therapies that can help their patients. Employing bureaucratic barriers to accessing these therapies is inappropriate.
Attached is a just completed meta-analysis on the subject of oncologist prescribing (Attachment B: “The Myth of Perverse Physician Incentives: Examining Research and Accusations in the Medicare Part B ASP Reimbursement System for Oncology”). It thoroughly debunks the false narrative that oncologists’ decision-making is driven by financial factors, much of which is based on extremely outdated and/or fundamentally flawed research. In fact, a number of studies on physician prescribing under the Part B reimbursement system have been published in recent years, all of which reach dramatically differing conclusions on physician prescribing patterns. They identify the following as drivers of drug selection by oncologists:
Concerns About the IPI Model as Proposed and Interactions with Other Models and Federal Programs
CMS requested comments on how to avoid unintended consequences of the interaction between the IPI Model as proposed and other federal programs, as well as the potential impact of the inclusion of manufacturer sales to model vendors on “best price” and average manufacturer price (“AMP”), and whether CMS should exempt prices offered under the model from AMP and “best price” calculations.
It is important that any Part B measures not disrupt or further complicate existing value-based care initiatives. While COA is totally committed to increasing the quality of cancer care and decreasing costs, as well as supporting value-based models that can positively impact prices and utilization, the IPI Model as proposed is not value-based. The IPI Model appears to be designed to target only costs and gives no consideration to how it will impact quality of care, access to treatment, or the utilization of those treatments.
COA is already working to spearhead innovative payment reforms aimed at lowering costs and improving the quality of cancer care, notably including the OCM. Both COA and CMS have invested heavily in the OCM and we urge CMS to consider the impact that the IPI Model as proposed would have on patients that are already benefitting from the OCM and other ongoing commercial value-based payment models. We believe that the threats to the OCM presented by the IPI Model could represent an unfortunate loss of years of effort, resources, and funds if the IPI Model as proposed is implemented.
Concerns with the Indexing of Part B Drug Reimbursement to International Prices
To reiterate, we are very concerned about the increasing cost of Part B drugs based on increasing prices. However, we do not believe that the international indexing component of the IPI Model is a realistic solution to decreasing Part B drug costs. We are concerned about the feasibility of indexing drugs to international prices, about the impact on all cancer drug reimbursement (Medicare fee-for-service, MA, and commercial), and about the operational implementation of index pricing.
In terms of the feasibility of indexing cancer drug prices to international prices, we are concerned about the consistent availability of cancer drugs across all countries to arrive at consistent pricing. In a number of the countries that would be involved in the indexing basket of drugs, the same drugs available in the U.S. are not available in those countries – or are available for a limited set of indications or are available under different names. Additionally, manufacturers may possibly increase international “list” prices while using rebates and discounts to bring down the true “net” price of a drug in other countries. As it is, we believe that U.S. “list” drug prices are forced higher by the increasing scope and magnitude of required rebates and discounts, the latter notably including discounts required under the 340B Drug Pricing Program (“340B”). The increasing scope and magnitude of 340B drug discounts are fueling “list” prices of Part B drugs. Equating U.S. “net” drug prices (i.e., ASP) with other countries, where the “list” price may not be reflective of rebates and discounts, creates “apples and oranges” pricing.
The problems just discussed may create a situation wherein the arrived at IPI Model index price is artificially and unrealistically low. Although at first blush this appears to benefit Medicare and its beneficiaries by lowering prices, this is a house of cards that is not realistic and sustainable. There is no way a CAP middleman vendor, or even a provider, will be able to use an artificially low international price to leverage that price, or lower, from a manufacturer, given that most oncology drugs do not have competitive substitutes (either generic, biosimilar, or brand therapeutic substitutes). What happens then? Will CMS reimburse the CAP vendor (or the provider, if not CAP vendor) for a drug at the index price, even though the vendor, or the provider, cannot purchase the drug anywhere close to that price? No CAP “vendor” will participate in the program under that scenario and forcing a provider to accept drug reimbursement for less than what a drug can be purchased for is simply not financially feasible.
Additionally, we note that any models involving ASP-based reimbursement must be exempted from manufacturers’ calculation of ASP so as to not inappropriately bring down the ASP calculation for those not participating in the proposed IPI Model, or other models. And CMS must not apply the 2% Medicare sequestration payment cut to any drug reimbursement models that are tested.
Finally, we note that “price fixing” of any goods and services has never worked in the long-term in the history of the world and typically leads to shortages, which already is a problem with low-priced generics. There are more effective and sustainable ways of lowering drug prices/costs through competition and market reform.
Comments on Alternatives to the IPI Model Components
As we have discussed, community oncologists are very concerned about access to and affordability of treatment for our patients, as well as the viability of our practices to continue providing the highest quality, most affordable, and most accessible cancer care. These concerns have guided us in analyzing the proposals in the IPI Model and in developing and analyzing possible alternatives. As we referenced earlier in this letter, COA is working on possible alternatives including the following:
We note, as COA has done on numerous occasions, that the add-on to ASP is not “profit,” as some falsely assert. It has to cover the very real increasing costs of human resources and infrastructure required to procure, handle, store, inventory, and dispose of chemotherapy and other cancer-related Part B drugs. What is left over covers operating expenses and bad debt. As it is, in the Medicare Physician Fee Schedule for 2019, community oncology practices will be reimbursed 3% less for administering chemotherapy. This is nonsensical and illustrates the problems of flat fees determined by Medicare formulas updated annually.
What COA is considering and analyzing, and recommends that CMS considers, is an alternative to a flat fee – a tiered percentage on ASP. More specifically, we are analyzing the following tiers of ASP add-on payments:
Note that we are not suggesting in any way that finances dictate oncologists’ choices in making treatment decisions in consultation with their patients. However, given both the escalating cost of brand drugs and what will be the introduction of more biosimilars in oncology, a tiered add-on will create a more appropriate reimbursement structure for what the add-on to ASP covers. Additionally, with biosimilars the goal is to create a robust and healthy market. With generics, increasing reimbursement will substantially help mitigate the constant shortages plaguing the market, as FDA Commissioner Scott Gottlieb, MD recently suggested.9
We also note that a tiered add-on to ASP, coupled with the clinically appropriate utilization management concept outlined below, will impact manufacturers’ pricing decisions for newly launched therapies depending on their degree of innovativeness. COA wants to maintain and reward drug therapy innovation and to continue to have access to new innovative therapies for our patients, but while controlling the costs of new therapies.
Our goals in considering this tiered ASP percentage add-on, in conjunction with employing clinically appropriate utilization management (as described below),
We realize that more detail is needed in this proposal, but the timing of this comment letter submission deadline only allows us to detail this proposal as much as we can at this point.
We believe that a CAUM provider-led model would be patient-centric, but at the same time, it would provide CMS an opportunity for savings by ensuring that Part B drug utilization best reflects value and efficacy, while encouraging more competition between manufacturers. Coupled with a tiered ASP add-on percentage, as outlined above, CAUM would create an environment that drives both manufacturer research and development, as well as oncologist decision making, in collaboration with patients, towards high value (a function of both quality and cost). Manufacturers would be driven by value in research and development, both in terms of therapy innovation and pricing. Oncologists and other Part B providers would be driven by the same value in adhering to provider-developed pathways, with Part B drug reimbursement equitable for generics, biosimilars, and brands. Finally, this would be in an environment devoid of any PBMs and related middlemen dictating treatment decisions based on rebates and other financial incentives feeding their bottom line profits rather than on what is clinically appropriate and in patients’ best interests.
Conclusion
As we have detailed very extensively in this letter and the supporting attachment, COA cannot support any mandatory experiment on patient care using CMMI as the IPI Model proposes. Rather, we hope to be able to continue working with CMS on constructing a patient-centric, value-driven, voluntary demonstration project with appropriate patient safeguards. COA would enthusiastically support this and help drive large-scale participation by community oncology practices across the country. The OCM provides the perfect model of collaboration whereby CMMI and providers have been working together to develop, implement, and refine a voluntary model of oncology payment reform.
Although COA has serious concerns with the IPI Model as proposed, we are very actively developing and analyzing alternative approaches. This should be demonstrated by our meetings and calls with CMS leadership and staff, as well as the comments provided in this letter. If you examine COA’s work over the past eight years, and the specific programs implemented by practices, community oncology has done more to advance payment reform in general and very specifically in cancer care. This dedication to doing what is right has come at a time when community oncology practices have faced declining Medicare reimbursement; enormous obstacles to providing clinically appropriate care, such as increasing inappropriate prior authorizations by insurers and other middlemen; and the destructive and harmful behavior of PBMs in obstructing cancer patients from getting timely and appropriate cancer care.
COA’s unwavering commitment and steadfast determination to continually improve our cancer care system is driven by a mission to ensure that patients with cancer, the majority of which are seniors who are Medicare beneficiaries, continue to have access to the highest quality, most affordable, and most accessible cancer care in the communities where they live and work.
COA looks forward to working closely with CMS to advance meaningful, patient-centered, and value-driven policies relating to cancer care. We are available to discuss any of our concerns and recommendations provided in this letter.
We thank you for your consideration.
Sincerely,
Jeff Vacirca, MD
President
Michael Diaz, MD
President Elect
Ted Okon
Executive Director
COA Comments on IPI for Medicare Part B Drugs
The Honorable Seema Verma, Administrator
Centers for Medicare & Medicaid Services
Department of Health and Human Services
Attention: CMS-5528-ANPRM
P.O. Box 8016
Baltimore, MD 21244-8013
Re: Medicare Program; International Pricing Index Model for Medicare Part B Drugs; CMS-5528- ANPRM
Dear Administrator Verma:
On behalf of the Board of Directors of the Community Oncology Alliance (COA), we are submitting this comment letter regarding the Medicare Program; International Pricing Index Model for Medicare Part B Drugs (CMS-5528-ANPRM) (the “IPI Model”).
As you know, COA is an organization that is dedicated to advocating for the complex care and access needs of cancer patients and the community oncology practices that serve them. COA is the only non-profit organization in the United States dedicated solely to independent community oncology practices, which serve the majority of Americans receiving treatment for cancer. COA’s mission is to ensure that cancer patients receive quality, affordable, and accessible cancer care in their own communities where they live and work. For more than 15 years, COA has built a national grassroots network of community oncology practices to advocate for public policies to support cancer patients.
First, we want to say that we appreciate your openness and interest in understanding our position on this important topic of the cost of prescription drugs in the Medicare Part B program (“Part B”), as well as the overall total cost of cancer care, which is comprised of drugs and services, both hospital and physician related. We thank you for meeting with us on two occasions leading up to the preparation and submission of this comment letter, as well as the opportunity to submit this letter. Although we are very concerned about escalating drug prices/costs and the high total cost of cancer care, as we have expressed to you, COA does not support the IPI Model as proposed in the pre- proposed rule published by the Centers for Medicare & Medicaid Services (“CMS”) because we have serious concerns about its impact on cancer patient care and even its legality. That said, we are actively working on solutions to high drug costs and the total cost of cancer care and, as we outline in this comment letter, are developing and analyzing alternatives to the model.
COA is sincerely committed to oncology payment reform that makes cancer care more affordable, as witnessed by our commitment and involvement in the Oncology Care Model (“OCM”); our ongoing development of the OCM 2.0, a next-generation, more universal oncology payment model; the numerous summits and meetings we have held with payers, employers, stakeholders, and providers; and the involvement of community practices, like ours, in an incredible number of private insurance payment models and programs. Community oncologists are concerned about the escalating prices and costs of cancer drugs, as well as high hospital-related costs, all of which contribute to the overall increasing total cost of cancer care. As leaders in the delivery of cancer care, we are mindful of our responsibility to be good stewards of costs we can control, including the utilization of drugs and services.
In this letter, we will detail our problems and concerns regarding the IPI Model as proposed, even though we have summarized these with you, leadership, and staff from CMS and the CMS Innovation Center (“CMMI”) in previous exchanges. We understand and appreciate that CMS appears to be already considering changes to its proposed model, and implementation of it, based on input from COA and providers. However, we believe it is important to detail our concerns with the IPI Model as proposed, for both the sake of memorializing our comments and as understanding the basis for our alternative suggestions.
In this comment letter, we will also outline alternatives to the IPI Model, which we are currently working on in greater detail. Some of that detail will not be available in this comment letter, given the submission deadline of 12/31 and the holiday season, coupled with the fact that we have much more analytical work to do on these alternatives.
We hope to continue the dialogue with CMS post submission of this comment letter, as we continue our analytical process and CMS moves towards a proposed rule on the IPI Model.
Summary Comments
COA does not support the IPI Model, and we strongly urge CMS not to move forward with it as proposed. We are greatly concerned that the IPI Model, which would conduct nothing short of a mandatory national experiment on Medicare Part B beneficiaries, could disrupt access to the innovative therapies and care that vulnerable seniors with cancer and other serious diseases need and are guaranteed under Medicare. The reasons for COA’s position are summarized as follows:
We also note, for the record, that COA is adamantly against any mandatory demonstration projects conducted by CMMI. As we detail in the attached legal analysis (Attachment A: Legal and Constitutional Problems with a Mandatory International Pricing Index Model), mandatory demonstration projects are clearly not in the charter of CMMI as written into law by Congress. If CMS believes that CMMI has the power via statute to effectively amend Medicare law – in this case the Part B drug reimbursement rate for at least 50% of Part B providers – it (CMS representing the Executive branch) either has overstepped its constitutional boundaries separating the powers of government branches or Congress has effectively handed over its powers to the Executive branch. That would either be illegal or unconstitutional, with the latter case invalidating the section of the law that created and funded CMMI.
By their nature, “mandatory” demonstrations are such because they force providers to participate and do not have stakeholder input or buy-in. This was the case with the attempt by the previous administration to force 75% of Part B providers to participate in the failed Medicare Part B Purchasing Model in late 2016.
Finally, we note two other concerns with the IPI Model as proposed. First, we have concerns about operationalizing an international pricing index as Part B drug reimbursement and about the adverse impact on commercial reimbursement, as further detailed in the aforementioned attached document (Attachment A). Second, we have several concerns about a “flat” average sales price (“ASP”) add-on, or any type of “capitated” fee payment, which we will also touch on later in this comment letter.
In summary, COA is very concerned that the IPI Model as proposed will totally upend how cancer care is delivered under Part B given the proposed mandatory national scope. The introduction of middlemen to Part B will seriously threaten the access of seniors and other Medicare beneficiaries to cancer treatment currently being delivered on an immediate “just in time” basis and close to home. Additionally, we are concerned with how such widespread changes (impacting 50% of Part B providers) to Part B reimbursement will impact the viability of community cancer clinics to continue providing care.
COA’s concerns about the IPI Model as proposed are based not on supposition or unfounded threats but on two clear facts. First, as we have documented in four volumes of actual patient cases1, PBMs and related middlemen are having an increasingly adverse effect on the treatment of cancer patients under Medicare Part D and commercial pharmacy benefit plans providing oral cancer drugs. Second, as we have documented along with others, there has been a marked shift of cancer care from physician-run community sites into large hospital systems due to declining reimbursement, especially in Part B, and clinic closings in rural areas. Data on this shift can be clearly seen in the annual COA Practice Impact Report, which tracks the changing landscape of community cancer care in the U.S. In the 2018 report, we note that 1,653 community oncology clinics and/or practices have closed, been acquired by hospitals, undergone corporate mergers, or reported that they are struggling financially. An average of 3.5 community oncology practices have closed per month, a rate that remains unchanged since the last report issued in 2016. Overall, 13.8 practices per month have closed, been acquired by hospitals, or undergone mergers since 2008.2 A recent feature in the Houston Chronicle newspaper makes this trend of closures real, telling the story of the anticipated closing of the Rio Bravo Cancer and Blood Clinic, which is currently providing cancer care to patients from over a 10,000 square mile area in rural Texas. As related in the story, “Most patients are seniors and dependent on Medicare…. But the clinic has been in the red for about 18 months because of a series of Medicare cuts for cancer drugs….”3
Radically changing Medicare reimbursement on a national basis, without studying its impact, could accelerate the shift in the site of cancer care from independent community cancer clinics to much more expensive hospital systems, resulting in higher costs for Medicare, commercial insurers, and beneficiaries, as well as in serious cancer care access issues, especially in rural or underserved areas.
As we have stated, COA appreciates that CMS is actively soliciting stakeholder input on what is a proposed model in a pre-proposed rule. Also, we are encouraged that CMS is listening to alternatives and other approaches.
In terms of specific alternatives that we are considering and analyzing, they include:
Comments on Concerns with the IPI Model as Proposed
Opposition to the Mandatory National Scope of the IPI Model as Proposed
The IPI Model as proposed is one of the largest and most complex CMMI demonstrations developed to date, with a broad geographic reach and strong potential to disrupt both the drug supply chain and the care delivery system. COA is staunchly opposed to mandatory models, especially of the scale and complexity outlined as part of the IPI Model. As previously expressed, we are against using mandatory demonstrations as a vehicle for the Executive branch to effectively bypass Congress by making sweeping changes to Medicare law. The IPI Model establishes a “demonstration” of the size and scope that far exceeds anything that can be reasonably considered a “test.”
COA’s legal and constitutional reasons for our opposition to mandatory CMMI demonstration projects are summarized as follows:
Attached (Attachment A) is a detailed overview of the legal and constitutional problems with the IPI Model as proposed.
COA is also very concerned that the mandatory nature of the IPI Model and the completely randomized process to divide the country into experimental and control groups would create insurmountable operational changes and would disadvantage patients based on geography. We do not believe that the agency can arbitrarily select geographies without creating significant disparities in access to care for seniors and other Medicare beneficiaries with cancer. Further, because the entire supply chain for oncology products would be disrupted, the IPI Model will impact providers both inside and outside the demonstration. As such, the IPI Model will impact not only 50% of Part B, but also Medicare Advantage (“MA”) and commercial plans and beneficiaries – in effect, the entire country. Additionally, this will present serious treatment dilemmas for practices, providers, and patients, and does not represent a solid experimental design. Any wholesale changes to Part B, especially impacting care delivery, should first be carefully tested on a limited scale, through programs that are voluntary for both patients and providers, in order to avoid creating operational complexities, increasing administrative burdens, and potentially even adding more costs to Medicare and its beneficiaries.
We believe that the IPI Model would hurt independent community oncology providers, who, ironically, do not represent significant cost burdens to the Part B program. According to the Medicare Payment Advisory Commission (“MedPAC”), Part B spending has been growing more rapidly for hospital outpatient departments (“HOPDs”) than for physicians and suppliers. Between 2009 and 2015, Part B drug spending grew at an average of roughly 16% for HOPDs and about 7% for physicians.5 Similarly, a recent peer-reviewed study found that both the mean total cost of care and the mean chemotherapy costs per patient per month were significantly lower for patients treated in a community-based practice compared with those treated in a hospital-based outpatient facility.6
We understand that the administration sees the increase in spending on oncology therapies as a challenge for patients and the federal budget. However, we do not agree that it is appropriate to address Part B cost growth by targeting community practices and independent physicians and endangering their ability to practice, especially given their proven role in driving value and efficiency in cancer care. Furthermore, cancer is a complex and sensitive therapeutic area and it does not lend itself well to models that seek to interfere with evidence-based clinical decision-making. Oncologists should not be punished for embracing medical innovation in the form of immunotherapies, targeted therapies, and personalized medicines that have transformed care and given patients with cancer a fighting chance.
Concerns about CAP and Middlemen as CAP Vendors in the IPI Model as Proposed
Under the IPI Model as proposed, CMS is looking to leverage its dormant authority under the previous failed CAP to contract with private-sector “vendors” to act as middlemen between manufacturers, physicians, HOPDs, and other providers for the drugs and biologic products included in the IPI Model.
COA strongly opposes any proposal to incorporate middlemen into the Part B program and empowering those entities to wield influence on cancer treatment. We firmly believe that all Medicare seniors, especially those facing life-threatening diagnoses like cancer, should have timely and unhindered access to the therapies that their physicians believe to be most effective for them. We do not believe that allowing corporate middlemen – who do not have in-depth clinical oncology experience – into Part B under CAP will improve patient care or outcomes. In fact, experience documents that it will do the exact opposite.
Under the IPI Model, physicians would be limited in their ability to treat patients to the best of their medical knowledge and experience, as CAP vendors would be able to directly or indirectly dictate the choice of therapies in the model. As cancer treatment becomes more precise and individualized, introducing an additional layer between patients and life- saving medicines is not only inappropriate, it could be very dangerous for seniors facing life-threatening diagnoses.
Furthermore, we are concerned that adding middlemen to Part B through the IPI Model would introduce the same cancer treatment access challenges experienced by cancer patients today in Part D. COA is befuddled by the administration’s interest in increasing the power and prevalence of middlemen at a time when both policymakers and the public have a heightened awareness of the business model and distorted incentives that have allowed PBMs to consolidate and grow their influence without reducing costs. In Medicare Part D, PBMs have increasingly sought to control costs by using a variety of mechanisms to steer patients toward less expensive therapies, but ultimately unduly complicating drug procurement, delaying and denying patients’ treatment, and driving up costs.
COA believes that the IPI Model could repeat past reform mistakes. The original CAP program, which ran for 18 months between 2006 and 2008, has been suspended for the last decade for good reasons. The original program presented a host of issues for Medicare, including negative impacts on access and supply chain dynamics and limited participation from both physicians and vendors. The original CAP program, we note, did not effectively lower costs for Medicare or for patients, and added significant complexities to drug negotiations and procurements. If CMS decides to reimpose CAP on the Medicare program, it will force the creation of individual patient inventories, increase the likelihood of treatment errors and delays, and place new and unreimbursed administrative burdens on clinicians. While the agency technically has the statutory authority to bring back CAP, COA strongly urges the administration to reconsider this proposal. CAP was not the answer to increasing competition in Part B in 2008, and it is not the answer in 2018.
Concerns About Included Drugs and Biologicals in the IPI Model as Proposed
It is unclear which drugs or therapeutic classes would ultimately be included under the proposed demonstration, but CMS has indicated its interest in including the “top spending” Part B drugs in the IPI Model. The agency has stated that it would likely start with an initial set of drugs in the early model years before expanding to a broader set of therapies.
COA believes that the IPI Model proposal should be withdrawn, but at a minimum cancer drugs should be excluded. Cancer treatment is a complex and sensitive therapeutic area, and it is inappropriate to subject patients with cancer and the providers that treat them to large-scale experimentation that could threaten patient outcomes.
Since the Secretary has the authority to exclude any drugs and biologicals whose inclusion is unlikely to result in cost savings or whose inclusion would have an adverse effect on access, we would like to flag some issues related to access to cancer drugs under the original CAP7 to point to the need to carve oncology out of any new CAP demonstration, regardless of how it is modified:
All Medicare seniors should have access to all therapies available, and the IPI Model as proposed would severely restrict the use of both new and standard-of-care cancer treatments. Cancer patients rely heavily on innovation in treatments and COA is concerned that every element of this model seriously challenges access to therapies and innovation. As the newest and innovative therapeutic products are, by definition, single-source, this proposal could limit the availability of the best available options in cancer treatment for seniors and other Medicare beneficiaries in the Part B program. Furthermore, CMS’ proposal could disincentivize lifesaving innovations.
COA believes the current Part B system offers the flexibility to use and administer the most clinically appropriate therapies based on clinical considerations, without delays or disruptions in cancer care.
Concerns About Flawed Notions of Part B Prescribing in the IPI Model as Proposed
The focus on changing the provider payment for drugs is grounded in the flawed notion that providers choose treatments based on financial considerations, rather than clinical ones. This is both incorrect and offensive. Independent, peer-reviewed research has found that modifying payment for cancer care does not impact clinical decision making for providers, but rather that the factors that actually impact prescribing patterns are the introduction of new drugs, new clinical evidence, and identifying new best-practices for treatment.8 In short, medical professionals make medical decisions using clinical evidence, not for financial gain. Providers already face the burdens associated with identifying and educating themselves on innovative techniques and therapies that can help their patients. Employing bureaucratic barriers to accessing these therapies is inappropriate.
Attached is a just completed meta-analysis on the subject of oncologist prescribing (Attachment B: “The Myth of Perverse Physician Incentives: Examining Research and Accusations in the Medicare Part B ASP Reimbursement System for Oncology”). It thoroughly debunks the false narrative that oncologists’ decision-making is driven by financial factors, much of which is based on extremely outdated and/or fundamentally flawed research. In fact, a number of studies on physician prescribing under the Part B reimbursement system have been published in recent years, all of which reach dramatically differing conclusions on physician prescribing patterns. They identify the following as drivers of drug selection by oncologists:
Concerns About the IPI Model as Proposed and Interactions with Other Models and Federal Programs
CMS requested comments on how to avoid unintended consequences of the interaction between the IPI Model as proposed and other federal programs, as well as the potential impact of the inclusion of manufacturer sales to model vendors on “best price” and average manufacturer price (“AMP”), and whether CMS should exempt prices offered under the model from AMP and “best price” calculations.
It is important that any Part B measures not disrupt or further complicate existing value-based care initiatives. While COA is totally committed to increasing the quality of cancer care and decreasing costs, as well as supporting value-based models that can positively impact prices and utilization, the IPI Model as proposed is not value-based. The IPI Model appears to be designed to target only costs and gives no consideration to how it will impact quality of care, access to treatment, or the utilization of those treatments.
COA is already working to spearhead innovative payment reforms aimed at lowering costs and improving the quality of cancer care, notably including the OCM. Both COA and CMS have invested heavily in the OCM and we urge CMS to consider the impact that the IPI Model as proposed would have on patients that are already benefitting from the OCM and other ongoing commercial value-based payment models. We believe that the threats to the OCM presented by the IPI Model could represent an unfortunate loss of years of effort, resources, and funds if the IPI Model as proposed is implemented.
Concerns with the Indexing of Part B Drug Reimbursement to International Prices
To reiterate, we are very concerned about the increasing cost of Part B drugs based on increasing prices. However, we do not believe that the international indexing component of the IPI Model is a realistic solution to decreasing Part B drug costs. We are concerned about the feasibility of indexing drugs to international prices, about the impact on all cancer drug reimbursement (Medicare fee-for-service, MA, and commercial), and about the operational implementation of index pricing.
In terms of the feasibility of indexing cancer drug prices to international prices, we are concerned about the consistent availability of cancer drugs across all countries to arrive at consistent pricing. In a number of the countries that would be involved in the indexing basket of drugs, the same drugs available in the U.S. are not available in those countries – or are available for a limited set of indications or are available under different names. Additionally, manufacturers may possibly increase international “list” prices while using rebates and discounts to bring down the true “net” price of a drug in other countries. As it is, we believe that U.S. “list” drug prices are forced higher by the increasing scope and magnitude of required rebates and discounts, the latter notably including discounts required under the 340B Drug Pricing Program (“340B”). The increasing scope and magnitude of 340B drug discounts are fueling “list” prices of Part B drugs. Equating U.S. “net” drug prices (i.e., ASP) with other countries, where the “list” price may not be reflective of rebates and discounts, creates “apples and oranges” pricing.
The problems just discussed may create a situation wherein the arrived at IPI Model index price is artificially and unrealistically low. Although at first blush this appears to benefit Medicare and its beneficiaries by lowering prices, this is a house of cards that is not realistic and sustainable. There is no way a CAP middleman vendor, or even a provider, will be able to use an artificially low international price to leverage that price, or lower, from a manufacturer, given that most oncology drugs do not have competitive substitutes (either generic, biosimilar, or brand therapeutic substitutes). What happens then? Will CMS reimburse the CAP vendor (or the provider, if not CAP vendor) for a drug at the index price, even though the vendor, or the provider, cannot purchase the drug anywhere close to that price? No CAP “vendor” will participate in the program under that scenario and forcing a provider to accept drug reimbursement for less than what a drug can be purchased for is simply not financially feasible.
Additionally, we note that any models involving ASP-based reimbursement must be exempted from manufacturers’ calculation of ASP so as to not inappropriately bring down the ASP calculation for those not participating in the proposed IPI Model, or other models. And CMS must not apply the 2% Medicare sequestration payment cut to any drug reimbursement models that are tested.
Finally, we note that “price fixing” of any goods and services has never worked in the long-term in the history of the world and typically leads to shortages, which already is a problem with low-priced generics. There are more effective and sustainable ways of lowering drug prices/costs through competition and market reform.
Comments on Alternatives to the IPI Model Components
As we have discussed, community oncologists are very concerned about access to and affordability of treatment for our patients, as well as the viability of our practices to continue providing the highest quality, most affordable, and most accessible cancer care. These concerns have guided us in analyzing the proposals in the IPI Model and in developing and analyzing possible alternatives. As we referenced earlier in this letter, COA is working on possible alternatives including the following:
We note, as COA has done on numerous occasions, that the add-on to ASP is not “profit,” as some falsely assert. It has to cover the very real increasing costs of human resources and infrastructure required to procure, handle, store, inventory, and dispose of chemotherapy and other cancer-related Part B drugs. What is left over covers operating expenses and bad debt. As it is, in the Medicare Physician Fee Schedule for 2019, community oncology practices will be reimbursed 3% less for administering chemotherapy. This is nonsensical and illustrates the problems of flat fees determined by Medicare formulas updated annually.
What COA is considering and analyzing, and recommends that CMS considers, is an alternative to a flat fee – a tiered percentage on ASP. More specifically, we are analyzing the following tiers of ASP add-on payments:
Note that we are not suggesting in any way that finances dictate oncologists’ choices in making treatment decisions in consultation with their patients. However, given both the escalating cost of brand drugs and what will be the introduction of more biosimilars in oncology, a tiered add-on will create a more appropriate reimbursement structure for what the add-on to ASP covers. Additionally, with biosimilars the goal is to create a robust and healthy market. With generics, increasing reimbursement will substantially help mitigate the constant shortages plaguing the market, as FDA Commissioner Scott Gottlieb, MD recently suggested.9
We also note that a tiered add-on to ASP, coupled with the clinically appropriate utilization management concept outlined below, will impact manufacturers’ pricing decisions for newly launched therapies depending on their degree of innovativeness. COA wants to maintain and reward drug therapy innovation and to continue to have access to new innovative therapies for our patients, but while controlling the costs of new therapies.
Our goals in considering this tiered ASP percentage add-on, in conjunction with employing clinically appropriate utilization management (as described below),
We realize that more detail is needed in this proposal, but the timing of this comment letter submission deadline only allows us to detail this proposal as much as we can at this point.
We believe that a CAUM provider-led model would be patient-centric, but at the same time, it would provide CMS an opportunity for savings by ensuring that Part B drug utilization best reflects value and efficacy, while encouraging more competition between manufacturers. Coupled with a tiered ASP add-on percentage, as outlined above, CAUM would create an environment that drives both manufacturer research and development, as well as oncologist decision making, in collaboration with patients, towards high value (a function of both quality and cost). Manufacturers would be driven by value in research and development, both in terms of therapy innovation and pricing. Oncologists and other Part B providers would be driven by the same value in adhering to provider-developed pathways, with Part B drug reimbursement equitable for generics, biosimilars, and brands. Finally, this would be in an environment devoid of any PBMs and related middlemen dictating treatment decisions based on rebates and other financial incentives feeding their bottom line profits rather than on what is clinically appropriate and in patients’ best interests.
Conclusion
As we have detailed very extensively in this letter and the supporting attachment, COA cannot support any mandatory experiment on patient care using CMMI as the IPI Model proposes. Rather, we hope to be able to continue working with CMS on constructing a patient-centric, value-driven, voluntary demonstration project with appropriate patient safeguards. COA would enthusiastically support this and help drive large-scale participation by community oncology practices across the country. The OCM provides the perfect model of collaboration whereby CMMI and providers have been working together to develop, implement, and refine a voluntary model of oncology payment reform.
Although COA has serious concerns with the IPI Model as proposed, we are very actively developing and analyzing alternative approaches. This should be demonstrated by our meetings and calls with CMS leadership and staff, as well as the comments provided in this letter. If you examine COA’s work over the past eight years, and the specific programs implemented by practices, community oncology has done more to advance payment reform in general and very specifically in cancer care. This dedication to doing what is right has come at a time when community oncology practices have faced declining Medicare reimbursement; enormous obstacles to providing clinically appropriate care, such as increasing inappropriate prior authorizations by insurers and other middlemen; and the destructive and harmful behavior of PBMs in obstructing cancer patients from getting timely and appropriate cancer care.
COA’s unwavering commitment and steadfast determination to continually improve our cancer care system is driven by a mission to ensure that patients with cancer, the majority of which are seniors who are Medicare beneficiaries, continue to have access to the highest quality, most affordable, and most accessible cancer care in the communities where they live and work.
COA looks forward to working closely with CMS to advance meaningful, patient-centered, and value-driven policies relating to cancer care. We are available to discuss any of our concerns and recommendations provided in this letter.
We thank you for your consideration.
Sincerely,
Jeff Vacirca, MD
President
Michael Diaz, MD
President Elect
Ted Okon
Executive Director