The value of open access and a patient centric approach to oral oncolytic utilization in the treatment of Chronic Myelogenous Leukemia: A U.S. perspective

ABSTRACT

Introduction: Since the introduction of tyrosine kinase inhibitors (TKIs), the treatment of patients with chronic myelogenous leukemia (CML) has resulted in significant improvement in patient survival but at a higher pharmaceutical cost to payers. The recent introduction of generic imatinib presents an opportunity to lower pharmacy costs within a population that is growing due to improved survival. Recent literature has focused on the likely benefits to payers of step therapy through generic imatinib.

Areas covered: This review provides a perspective that is broader than the evaluation of financial savings or narrowly defined health economic metrics by incorporating factors such as CML patient heterogeneity, including varying levels of disease progression risk, comorbidities and genetic mutation status, differences in TKI product profiles, clinical guideline recommendations, and the importance of individualized patient care. A focused literature review evaluating the real-world impact of utilization management programs is presented.

Expert commentary: The findings indicate that payers can achieve substantial savings without the need to implement utilization management policies. Compromises in the ability to provide individualized patient care and unwanted economic consequences resulting from increased costs of disease progression, adverse events, and lack of response to treatment due to utilization management are summarized.

KEYWORDS:

• Chronic myelogenous leukemia
• open access
• step therapy
• imatinib
• dasatinib
• nilotinib
• tyrosine kinase inhibitor
• formulary restriction

1. Introduction

Chronic myelogenous leukemia (CML) is a progressive cancer of bone marrow stem cells representing 0.5% of all new cancer cases [1]. Surveillance, Epidemiology, and End Results Program of the National Cancer Institute (SEER) estimated that there were 8,220 new cases of CML in 2016 in the USA [1]. However, based on other estimates, this number can be much lower with estimates of 4,800–5,200 new cases [2]. The incidence of CML has remained relatively stable over time, while the prevalence has continued to increase due to improvements in survival. The prevalence of CML in the USA is estimated to more than double in 20 years with 70,000 cases in 2010 and a projected increase to 112,000 in 2020 and 144,000 in 2030 [2].

CML is characterized by an excess proliferation of hematopoietic stem cells triggered by the Philadelphia chromosome that results from a reciprocal translocation of the ABL gene from chromosome 9 onto the BCR gene on chromosome 22 [3]. Most patients are diagnosed in the chronic phase and both untreated and treated patients can progress to an accelerated phase (AP) and the fatal blast crisis (BC) phase [3,4]. The introduction of tyrosine kinase inhibitors (TKIs) has improved the CML treatment landscape beyond traditional cytotoxic chemotherapeutic agents or interferon-alpha. The first-generation TKI, imatinib, was approved in 2001 for first-line use. The second-generation TKIs (2G TKIs), dasatinib and nilotinib, were initially approved for second-line use in 2006–2007 and subsequently received a new indication for first-line use in 2011. Two other TKIs entered the market in the subsequent years, bosutinib and ponatinib; these potent treatments are commonly reserved in clinical practice for when other TKIs cannot be used (e.g. due to genetic mutations). These therapeutic advances have increased the average life expectancy to the levels comparable to the general population with steady improvements in the 5-year relative survival rates from around 50% to 90% over time [5]. Based on current clinical practice, TKIs are recommended as a lifelong treatment of CML [3]. Thus, it is crucial to understand the clinical and economic impact of any policies that could potentially limit patient access to treatment options.

The Gleevec® loss of exclusivity (LOE) in the USA brought about the entry of the first generic imatinib in February 2016 and with it, extensive attention from both payers and researchers due to potential cost savings that could be achieved through generic substitution of the branded agent [6–8] or by instituting utilization management policies like step edits and/or prior authorizations [6,7]. Generic substitution of a branded agent is most commonly a prescriber- and pharmacist-driven process that is governed by a combination of federal and state mandates. Utilization management, on the other hand, is a payer-driven process with the direct potential to impact treatment decisions and subsequent health outcomes.

The aim of this review was to evaluate the value of open access to TKIs in CML factoring in clinical and economic considerations from multiple perspectives including patients, providers, and payers. The likely approaches that payers may implement to control access to TKIs include two common utilization management techniques: (1) a generic imatinib step edit that requires newly diagnosed patients to fail generic imatinib prior to accessing branded 2G TKIs, and/or (2) a preferential formulary placement of one 2G TKI.

2. Generic imatinib step edit

There are clinical and economic limitations associated with implementing generic step edits. In CML patients, it is important to achieve disease control to reduce the risk of disease progression. The risk of disease progression can be measured using the Sokal [9], Hasford [10], or European Treatment Outcome Study (EUTOS) [11] scoring systems; Sokal and Hasford scales categorize risk as low, intermediate, or high, whereas the EUTOS scale has only two categories – low and high. The updated 2016 National Comprehensive Cancer Network (NCCN) guidelines state that 2G TKIs (dasatinib or nilotinib) are preferred over imatinib as first-line therapy for intermediate to high-risk patients [12,13]. There are several studies that provide estimates of the underlying distribution of disease progression risk scores in newly diagnosed CML patients. Clinical trials and real-world registry data have consistently shown that approximately two-thirds of patients are at intermediate or high risk of disease progression at diagnosis (Figure 1) [14–16]. Implementing generic imatinib step edits irrespective of disease progression risk level is inconsistent with the NCCN guidelines and has the potential to compromise clinical outcomes of a large proportion of CML patients.

Figure 1. Distribution of Risk Scores in Newly Diagonsed CML Patients [9,10,14–16].​

Note: Data from the DASISION trial is based on summing sample sizes across both treatment arms (dasatinib 100 mg = 259; imatinib 400 mg = 260) while the ENESTnd trial is based on summing sample sizes across all three treatment arms (nilotinib 300 mg = 282; nilotinib 400 mg = 281; imatinib 400 mg = 283). For the real-world registry data, the distribution of risk scores is based on all patients excluding those with missing values. Sokal risk groups: low-risk (Sokal score < 0.8), intermediate-risk (Sokal score 0.8–1.2) and high-risk (> 1.2). The Hasford score: low-risk (score ≤ 780), intermediate-risk (score 781 – 1480) and high-risk (score ≥ 1481).

Studies of imatinib therapy have demonstrated long-term efficacy [17,18]. One of these studies [17] enrolled approximately 88% low-risk patients. In another long-term follow-up study [18], outcomes were consistently inferior for high-risk patients. Studies evaluating imatinib dose escalation showed that high-dose imatinib improved the outcomes of CCyR and MMR [12] [19], but did not show improvements in rates of disease progression, compared to standard dose [12]. Moreover, high-dose imatinib was associated with higher rates of treatment interruptions due to toxicity [12].

The need to further improve clinical outcomes in CML led to the advent of 2G TKIs. There has been no incremental survival benefit shown to date for 2G TKIs versus imatinib. However, due to dramatically long survival of imatinib patients, definitive evidence on this outcome requires a follow-up that is longer than currently available. Meanwhile, clinical trials have consistently demonstrated significantly higher rates of response and lower rates of progression in patients treated with 2G TKIs [14,15] and benefits have persisted over a 5-year follow-up period [20,21]. Thus, long-term efficacy data indicate that: (1) 2G TKIs are significantly more efficacious relative to imatinib in achieving faster and deeper treatment responses; (2) 2G TKIs have a long-lasting response over time. The literature also indicates that the achievement of deep molecular response is associated with significant improvements in progression-free, event-free, and overall survival [22–25]. The need to more aggressively manage the disease in intermediate- and high-risk CML patients led to the NCCN treatment recommendation to prefer 2G TKIs over imatinib in 1st line setting. The introduction of generic imatinib should not change the appropriate management of CML.

As discussed earlier, 2G TKIs have been shown to be associated with lower rates of disease progression compared to imatinib. The NCCN recommended interventions for patients with advanced phase CML include escalation of TKI doses, induction chemotherapy, and a hematopoietic stem cell transplant (HSCT) if feasible [3]. NCCN guidelines do not recommend using high-dose imatinib for initial therapy due to high levels of toxicity associated with this dose, and recommend switching to a superior efficacy 2G TKI upon failing standard dose imatinib and reserves the use of high-dose imatinib only for advanced cases [3]. Allogeneic HSCT is preferred over the alternative, autologous HSCT, due to avoided risk of harvesting leukemia cells and the potential graft-leukemia effect [26]. The cost of care for patients experiencing progression can be very high. Specifically, for HSCT alone, a longitudinal study of private claims data from a cohort of 3,365 HSCT recipients between 2007 and 2009 identified a median 100-day total estimated cost of $99,899 for autologous HSCT and $203,026 for allogeneic HSCT [27]; these findings are similar to those reported by another study examining economics of HSCT [28]. Further, a recent study found that CML patients with evidence of disease progression incurred $270,925 more in one year total healthcare costs than those without disease progression (95% CI: $235,290, $311,958). After controlling for costs prior to progression, the incremental increase in 1-year costs was $136,308 (p < 0.0001) for patients with vs. without disease progression [29]. Thus, there is a substantial economic burden of progression.

From a traditional cost-effectiveness standpoint, stepping through generic imatinib may be more attractive than open access to 2G TKIs, assuming generic imatinib is clinically equivalent to branded imatinib, while being substantially cheaper. Sacha et al. demonstrated non-inferiority of generic imatinib in clinical efficacy and tolerability compared to branded imatinib in Poland at a significantly lower cost [30]. Padula et al. [6] used Markov models to compare the 5-year cost-effectiveness of an ‘imatinib first’ (i.e. step-edit) versus ‘physician’s choice’ that allowed for the initiation of any one of the three TKIs approved for first-line use, in achieving complete cytogenetic response (CCyR) and early molecular response (EMR). The findings from this study indicate that both approaches – imatinib first (CCyR: $71,679/quality adjusted life year (QALY); EMR: $73,588/QALY) and physician’s choice (CCyR: $92,135/QALY; EMR: $92,166/QALY) met the cost-effectiveness threshold of $100,000/QALY compared to a no-alternative-treatment option. In this model, step-edit patients incurred less overall cost over a 5-year period while achieving only slightly fewer quality-adjusted life benefits over the physician’s choice arm.

These findings should be interpreted keeping in mind several limitations. First, the model allowed only one opportunity to switch within the first 12 months after treatment initiation during a 5-year time horizon [6]. NCCN guidelines recommend regular evaluations of the response to treatment every 3–6 months depending on time on therapy. Lack of response may develop beyond the first 12 months of treatment due to multiple reasons, for example resistance to treatment caused by genetic mutations or lack of treatment adherence caused by tolerability issues, all of which would require a switch of therapy. Underestimating the switch rates from a low-cost generic imatinib likely resulted in overestimating the savings. Next, contrary to the NCCN treatment guidelines, under the ‘imatinib first’ scenario the model allowed to treat all newly diagnosed CML-CP patients with imatinib regardless of any patient factors, including the risk of disease progression. Further, the model also assumed that generic imatinib would be 60–80% of branded treatment cost for the first 6 months after the Gleevec LOE (180-day exclusivity), and 10–30% of branded treatment cost was assumed thereafter. These estimates do not correspond to the current pricing trends: the actual cost of generic imatinib for the first 6 months of Gleevec LOE was 94% of branded treatment cost and currently ranges from 86% to 94% of branded treatment [31].

These limitations and assumptions likely led to overestimating the savings under the imatinib first scenario. Finally, the patient population analyzed in the study included only newly diagnosed CML-CP patients. Given relatively low incidence of CML and long survival, newly diagnosed patients represent a small proportion of the total CML population [1,2]. Imatinib step therapy would be applicable only to this fraction of the total CML population who are new to treatment. Based on the finding that the estimated savings from imatinib-first versus physician’s choice were $80,000 to $90,000 per patient over 5 years, and in light of the fact that the model was developed for newly diagnosed CML patients, it does not reflect the proportion of the total TKI budget for the total CML population this saving represents.

A model was developed by Bloudek et al. [7] to estimate the economic impact of the entry of generic imatinib and potential formulary management to require generic imatinib step therapy before use of a branded TKI among patients new to TKI therapy. Findings indicated relatively minimal incremental cost savings from step edits in newly treated patients affected by generic step because of the small proportion of these patients relative to the total patient population [7]. The Gleevec LOE was expected to reduce total pharmacy spend on TKIs by nearly one-third over 2 years, independent of additional formulary management, through generic substitution of Gleevec, which had the largest market share prior to the LOE. Assuming generic step therapy was applied only to patients new to TKI therapy and that 74% of prior authorizations for branded TKIs would be approved, formulary management reduced TKI spending by an incremental 1.0–2.1%. In other words, nearly all of the budget impact of generic imatinib entry to payers was estimated to be realized without formulary restrictions.

In a CML treatment landscape that has been traditionally dominated by branded imatinib [32,33], it is clear that the introduction of generic imatinib shows tremendous potential for payer cost savings from generic substitution alone, without the need to implement any utilization management policies. The cost savings from generic imatinib by payers are not necessarily translated into savings for the patients. Ironically, generic substitutions might not be patient centric for certain patients like those enrolled in Medicare Part D programs, who may in fact, face higher out-of-pocket costs for generic medications. Even with a lower drug acquisition cost, this counterintuitive phenomenon occurs because manufacturers of branded drugs are required to take on a greater proportion of the ‘coverage gap’ (or donut hole), whereas manufacturers of generics are not required to provide such contributions, resulting in a higher out-of-pocket (OOP) expenditure for the patient while they are in the coverage gap [34].

The Affordable Care Act includes a provision to reduce the cost sharing to 25% by 2020, but until then CML patients may incur higher OOP costs for generics, as high as 51%, compared to 40% cost sharing for branded drugs while in the coverage gap phase of the Medicare Part D benefit design for 2017 [34]. Using the real drug acquisition costs from Analy$ource® First Databank for a 400 mg daily dose of imatinib generic versus Gleevec may result in significantly greater out of pocket burden for patients albeit this is likely not the only Part D therapy a CML patient may be receiving (Figure 2) [31]. Thus, payers, but not necessarily patients, will benefit from the cost savings of generic imatinib substitutions. Based on the list price current as of the date of this review, the savings from the generic in the catastrophic phase were still not large enough to result in overall annual savings to the patient. Given the fact that many Medicare Part D patients may be on fixed incomes, it is important to consider the potential impact on OOP costs and subsequent effects on adherence. Furthermore, patients on branded agents might be able to take advantage of co-pay assistance programs, more commonly offered by manufacturers of branded products than generic manufacturers, to lessen the OOP cost burden.

Figure 2. Annual Patient Out-of-Pocket Burden in 2017 using WAC List Prices [24].

Note: WAC, wholesale acquisition cost, as of August 2016; Analy$ource® First Databank. Imatinib NDC 47,335–0475-83; GLEEVEC® NDC 00078–0649-30. Based on 2017 Standard Benefit Design including $400 annual deductible, $3,700 initial coverage limit, and $4,950 in True Out of Pocket prior to entering catastrophic coverage.

3. Preferential formulary placement of one 2G TKI

Preferential formulary placement of either nilotinib or dasatinib is associated with several limitations. There are differences in the safety profiles of dasatinib and nilotinib. The most common adverse events associated with dasatinib are fluid retention events such as, pleural effusion, pericardial effusion, pulmonary arterial hypertension, and pulmonary edema [21,35]. Nilotinib-related adverse events include coronary artery stenosis, femoral arterial stenosis, peripheral arterial occlusive disease, and intermittent claudication [35,36]. Additionally, new evidence [37] is emerging for differences in the metabolic profiles between dasatinib and nilotinib. Treatment with nilotinib was associated with a worsening of the glycometabolic profile compared to patients treated with dasatinib and imatinib [37]. Further, claims data analysis from U.S. commercial and Medicare payers indicated that patients receiving nilotinib, either as first- or second-line therapy, had a higher risk of developing metabolic diseases like Type 2 diabetes mellitus [38] and hyperlipidemia [39] compared to those on dasatinib. Based on the differences in the safety profiles, the NCCN guidelines recommend consideration of certain comorbidities including heart disease, arrhythmia, diabetes, pancreatitis, pleural effusion, and lung disease while making treatment decisions related to 2G TKIs [3]. Real-world data indicated that approximately 41% of newly treated CML patients had at least one major comorbidity relevant to the selection of 2G TKI therapy for individual patients [40]. Specifically, CML patients had a high prevalence of comorbidities like heart disease (23%), diabetes (18%), lung disease (13%), and arrhythmias (7%) [40].

Cardiovascular disease (CVD), which appears to be the most common comorbidity among CML patients, may also have a higher prevalence among CML compared to the general population. The age- and gender-standardized prevalence rates of myocardial infarction, heart failure, atherosclerosis, and stroke and risk of CVD have been shown to be significantly higher relative to the general US population [41]. Furthermore, this increased risk of CVD is not explained by residence in the geographic regions of the USA with higher than average incidence of CVD [42].

Heterogeneity of patients’ comorbidity profile is reflected in the NCCN guidelines recommending that dasatinib is a preferred treatment in patients with a history of arrhythmias, heart disease, pancreatitis, or hyperglycemia, while nilotinib is a preferred treatment for patients with a history of lung disease or pleural effusion [3]. In addition to comorbidities, BCR-ABL1 mutations must be considered in TKI treatment selection. These mutations may develop any time during TKI treatment and impact the patient responsiveness to certain TKIs. Open access to all TKIs is critical at all stages of treatment to allow for customization of treatment depending on the mutational status. For example, genetic mutations that reduce responsiveness to one therapy are often different from those that reduce responsiveness to another one of the TKIs [43]. Thus, preferential formulary placement of one 2G TKI vs. another may be limited by patient heterogeneity in terms of comorbidity and genetic mutation profiles.

Differences in safety profiles of 2G TKIs and failing to individualize care may have economic consequences as well. Adverse events of TKIs may be very costly. A retrospective analysis [44] evaluated healthcare costs associated with adverse events among TKI-treated patients identified in the FDA Adverse Event Reporting System [35] that were potentially associated with TKIs. The study found (Figure 3) [44] that one year per patient total mean medical costs for these adverse events were $17,015 (femoral arterial stenosis, FAS), 15,145 (peripheral arterial occlusive disease, PAOD), $6,112 (intermittent claudication, IC), $4,944 (coronary arterial stenosis, CAS), $2,797 (pericardial effusion) and $1,908 (pleural effusion) [44]. Adverse events like FAS, PAOD, IC, and CAS are commonly associated with nilotinib, whereas, pericardial and pleural effusions are more commonly associated with dasatinib [35].

Figure 3. Total Medical Costs for Patients with Select TKI-related Adverse Events [37].

Note: FAS, femoral arterial stenosis; PAOD, peripheral arterial occlusive disease; Inter Claud, intermittent claudication; CAS, coronary arterial stenosis; Peri Eff, pericardial effusion; Pleu Eff, pleural effusion; CH, conjunctival haemorrhage

Additionally, lack of individualized care may lead to economic waste due to treatment resistance to TKIs caused by genetic mutations. Jabbour et al. [43] assessed the treatment response and economic outcomes associated with restricted access to 2G TKIs. This study [43] mimicked real-world second-line treatment considerations by incorporating data on the frequencies of the genetic mutations that occur in response to first-line treatments and the impact of these mutation on patient responsiveness to 2G TKIs like dasatinib and nilotinib. This study modeled 3 different scenarios – (1) plan limiting open access to both 2G TKIs, (2) plan limiting access to dasatinib only, and (3) plan limiting access to nilotinib only. The findings indicate that the highest CHR and MCyR rates and lowest costs per each of these responses were achieved in the setting with an open access to both of the first-line 2G TKIs (Figure 4) [43]. Ninety-three percent of patients achieved CHR in an open access plan compared to 88% and 67% of patients in dasatinib only and nilotinib only plans, respectively. Similarly, highest rate of MCyR was observed in the open access plan (56% open access, 53% dasatinib only, 47 % nilotinib only). Compared to the open access plan, the health plan that provided access to dasatinib alone cost 5% more and the plan that provided access to nilotinib alone cost 41% more per CHR event; restricted access plans costed 6% (dasatinib only) to 22% (nilotinib only) more per MCyR event. These findings underscore the complex and ever changing nature of the disease, as CML patients experience ongoing genetic mutations. Hence, it is critical to have an open access to a variety of TKI treatment options in order to provide the most optimized and individualized care for CML patients with a wide range of comorbidity and genetic profiles.

Figure 4. Total drug cost per complete hematologic- and major cytogenetic response in open access vs. restricted access settings [36].

Note: CHR, complete hematologic response; MCyR, major cytogenetic response; ?, difference

4. Real-world impact of utilization management policies

It is estimated that approximately 28% of health plans utilized step therapy policies in 2005 and the number has increased to 69% in 2014 [45]. The expansion of utilization management techniques into complex disease areas like oncology has resulted in concerns regarding the suitability and policy impact of these techniques [46]. Additionally, the efficiency of these techniques in one therapeutic area might not translate well into other disease areas due to inherent differences in the diseases, available therapeutic options, and the heterogeneity in their implementation [47]. The primary metric of success for these policies remains cost-containment from the payer perspective; the impact of these policies on patient outcomes, quality of care, or provider burden appears to be an afterthought and thus is less commonly evaluated [48]. There is consistent evidence from several reviews [49–52] suggesting positive short-term impact of utilization management policies on prescription drug related spending by payers [49–52]. These reviews also demonstrate an association between utilization management and reduced treatment adherence [51,52].

It remains unclear whether utilization management programs can reduce disease related costs, overall health care costs, or prevent cost shifting from pharmacy benefit to medical benefits, as the success of these programs has not been comprehensively evaluated [47,50,52]. There is consensus in the literature around the need for long-term research [47,49,50] from a variety of perspectives (payer, patient, provider) as well as including a set of relevant and comprehensive outcomes (clinical, economic, and quality) [49].

As it applies to CML, there may be multiple reasons for utilization management to potentially increase downstream costs. For example, as a result of formulary restrictions, patients that are ‘stepped’ through imatinib may have resistance to treatment that may not be detected due to poor molecular monitoring rates in the real world practice. The NCCN guidelines recommend regular molecular monitoring to determine treatment response: every 3 months after TKI initiation for 2 years and every 3–6 months in the following years after CCyR achievement. However, a recent analysis of US insurance claims data from 1,431 newly diagnosed patients indicated that 36% of patients did not receive any molecular tests and the average test frequency was 1.6 during the first year of CML treatment [53]. Insufficient monitoring may have a negative impact on clinical outcomes, lead to omission of treatment resistance, and consequently increase healthcare costs.

5. Conclusion

This review indicates that the primary driver of utilization management programs is focused on cost containment for health plans. From a payer perspective, formulary restrictions are likely to only offer limited and siloed financial benefits to payers while having negative consequences for providers and patients. Providing unrestricted access to TKIs for clinicians and patients would allow them to individualize care based on the risk of disease progression, comorbidities, and genetic mutations. The heterogeneity factors make no one TKI suitable for all CML patients. Current clinical guidelines reflect the inherent differences between TKIs and have accordingly recommended a variety of options.

6. Expert commentary

While compromising the ability to provide high-quality and individualized patient care, utilization management techniques may result in unwanted economic consequences resulting from increased costs of disease progression, adverse events, and lack of response to treatment. Evaluation of techniques currently proposed in the literature like increased enterprise-wide clinical collaboration, integration of pharmacy and medical management processes, improved care coordination and innovative treatment pathway development should be implemented in place of utilization management as they can deliver high-quality individualized care while controlling cost [54]. This review provides a perspective that is broader than the evaluation of financial savings or narrowly defined health economic metrics.

7. Five-year view

The future of utilization management within the oncology space is likely to continue to expand as new treatment options become available and cost of therapy continues to rise despite uncertain outcomes and newer alternative techniques. The passage of the 21st Century Cures Act has made available $1.8 billion for the development of oncology products and is expected to accelerate clinical innovation as well as provide quicker access to life-saving drugs through faster FDA approval processes. Clinical innovation in CML treatment has resulted in significant improvements in survival, resulting in increased prevalence and further financial burden to payers. It is envisioned that healthcare delivery will continue to become increasingly more integrated allowing for more balanced and comprehensive research across the spectrum of outcomes and spending with more definitive evidence that can lead to more effective management techniques. Emerging management approaches are designed to focus on quality of care as well as individualized care. Care models incorporating capitation, case mix, bundled payments, and pay-for-performance mechanisms that allow patients and providers to share the financial risks with payers while maintaining access to individualized patient centric care are expected to be the future of the US healthcare.

Key issues

• CML has increasing prevalence, stable incidence, and under the current open access scenario, TKI treated patients have life expectancies comparable to the general population

• CML is a complex disease with substantial patient heterogeneity where treatment decisions should depend on patient prognostic risk, presence of certain genetic mutations, comorbidities and consideration of risk factors for TKI related adverse events; response to treatment needs to be regularly monitored and treatment optimized in order to achieve best disease control and improve outcomes

• The expiration of the Gleevec patent may prompt payers to consider generic imatinib step edits in an attempt to control pharmaceutical spending in this area. However, payers are to achieve considerable savings through generic substitution without instituting utilization management policies

• Utilization management efforts are likely to interfere with guideline recommended treatment selection and result in poorer clinical outcomes that could potentially offset some of the cost savings achieved through formulary restrictions

• There is little to no evidence that utilization management works in a complex disease area like CML and instituting such policies is not recommended until supporting evidence is available

Declaration of interest

L Das and M Gitlin are consultants for BMS. LR Siegartel is a full-time consultant to BMS. D Makenbaeva is an employee of BMS and may own BMS stock. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

This manuscript was made possible due to funding from Bristol-Myers Squibb Company.