Real-world use and outcomes of hypomethylating agent therapy in higher-risk myelodysplastic syndromes: why are we not achieving the promise of clinical trials?
Abstract
Myelodysplastic syndromes are hematological malignancies characterized by ineffective hematopoiesis and a high risk of progression to acute myeloid leukemia. Hypomethylating agents (HMAs), azacitidine and decitabine, are standard of care therapy for higher-risk myelodysplastic syndromes. However, outcomes reported for real-world studies fall short of those achieved in clinical trials. We conducted a targeted literature review exploring real-world utilization, persistence and outcomes with intravenous and subcutaneous HMA therapies to better understand barriers to achieving optimal outcomes in clinical practice. The potential benefits of oral HMA therapy were also explored. Underutilization and poor persistence with HMA therapy are associated with suboptimal outcomes, highlighting the need for approaches to improve utilization and persistence, so that patients achieve the optimum benefit from HMA therapy.
Lay abstract
Myelodysplastic syndromes (MDS) are bone marrow disorders affecting the production of blood cells. In some patients, MDS can progress to acute myeloid leukemia (AML), an aggressive blood cancer with poor prognosis. Patients with higher-risk MDS are often treated with a type of chemotherapy called hypomethylating agents (HMAs). Studies conducted in real-world clinical practice have shown HMAs to be less effective than has been found in clinical trials. We reviewed available studies exploring real-world utilization, persistence and outcomes with current HMA therapies to better understand any barriers to patients achieving the best outcomes. Two important factors were found to be the underuse of HMAs and poor persistence with HMA therapy, highlighting the need for approaches to improve HMA utilization and persistence.
Myelodysplastic syndromes (MDS) are a heterogeneous group of hematological malignancies characterized by ineffective hematopoiesis. This manifests as cytopenias leading to anemia and the need for frequent red blood cell (RBC) transfusions, an increased risk of infections, and bruising and bleeding [1,2]. Patients frequently experience debilitating fatigue and shortness of breath. Particularly in those patients who are transfusion dependent, recurrent infections and significant bleeding events often necessitate emergency room (ER) visits and hospitalization [3]. Furthermore, these patients are at high risk of progression to acute myeloid leukemia (AML), for which the prognosis is very poor [1,2].
In 2017, the age-adjusted incidence rate for MDS in the USA was estimated at 3.7 per 100,000 people, and an estimated 58,471 patients were living with MDS [4]. Most patients with MDS are older adults, with a median age at diagnosis of 77 years, and incidence increases dramatically from the age of 70 years [5]. MDS is divided into risk categories based on disease- and patient-specific factors, which predict survival and the rate of progression to AML [6–9], and shape treatment throughout the patient’s disease course [10]. Overall survival (OS) is poor, especially for patients with intermediate- or high-risk disease, comprising around 43% of the MDS patient population (20% intermediate, 13% high and 10% very high risk) [7]. Median OS ranges from 3.0 years for intermediate-risk, to 1.6 years for high-risk and 0.8 years for very high-risk disease, according to the Revised International Prognostic Scoring System risk categories [7]. Most patients die from complications related to MDS. A large retrospective study observed that progression to AML accounted for almost half of deaths (47%), while infection (27%) and bleeding complications (10%) were other main causes of death [11].
Treatment of higher-risk MDS aims to modify the course of the disease [12]. The only curative therapy for MDS is allogeneic stem cell transplantation. However, few patients are considered eligible and undergo this procedure, given the high risk of procedure-related toxicities, especially in older patients [13]. The only alternative disease-modifying therapies are the hypomethylating agents (HMAs) azacitidine and decitabine (approved by the US FDA in 2004 and 2006, respectively, and by Health Canada in 2009 and 2019, respectively; azacitidine was authorized by the EMA for the treatment of MDS in 2008, and decitabine is approved by the EMA for AML), and the immunomodulatory agent lenalidomide, although the latter is only approved for the subgroup of patients with the deletion 5q mutation [12–14]. Other therapeutic options used in the management of certain patients with MDS include erythropoiesis-stimulating agents, luspatercept, low-intensity chemotherapy or biologic response modulators and immunosuppressive therapy; however, these are not typically considered disease modifying [13]. In clinical trials, HMAs have been shown to be effective for improving outcomes such as transfusion dependence and health-related quality of life (HRQoL), as well as delaying progression to AML [15–17]. Furthermore, azacitidine has been shown to significantly prolong OS compared with conventional/supportive care [17]. Reflecting these data, HMA therapies are now the standard of care for management of higher-risk MDS [12,13]. However, real-world outcomes among patients on HMA therapy have been shown to be much lower than those achieved in clinical trials [18–22], thus suggesting that there are shortcomings in the use of HMA therapy in routine clinical practice.
Until recently, HMA therapies were available only in intravenous (IV) or subcutaneous (SC) formulations, requiring administration in a hospital or other healthcare facility over several days. For example, azacitidine requires administration for 5–7 consecutive days by IV or SC injection in cycles of 4–6 weeks [23], while decitabine can be given as an IV infusion over 3 h repeated every 8 h for 3 days every 6 weeks, or as an IV infusion over 1 h for 5 days every 4 weeks [24]. These regimens may pose logistical challenges for patients and their caregivers, especially in the elderly, and may be a factor in the poorer outcomes of HMA therapy in the real-world versus clinical trials setting. To investigate this, we conducted a targeted review of the literature exploring real-world utilization, persistence and outcomes with IV and SC HMA therapies, so as to better understand barriers to achieving optimal outcomes in clinical practice. The potential benefits of oral therapy as an alternative to IV and SC HMAs in the management of higher-risk MDS were also explored.
Methods
This literature review was conducted using targeted search methodology to identify articles relevant to the review objectives, as described above. Initial searches were conducted in MEDLINE® (via PubMed®) in July 2020 for articles published in the prior 5 years, using key terms including, but not limited to, persistence, adherence, treatment patterns, real-world outcomes, oncology, cancer, myelodysplastic syndromes, hypomethylating agents, oral, intravenous, subcutaneous, decitabine and azacitidine. Reference lists of identified articles were reviewed and additional manual searches conducted for relevant studies not identified in the MEDLINE database search, including congress abstracts. All study designs were considered for inclusion.
Real-world utilization of HMA therapy for higher-risk MDS
Data from real-world studies suggest that around half of patients with higher-risk MDS do not receive HMA therapy (Table 1). For example, a retrospective cohort study compared treatment and outcomes for patients aged ≥66 years included in the Surveillance, Epidemiology and End Results (SEER)-Medicare database and diagnosed with refractory anemia with excess blasts (RAEB), used as a proxy for higher-risk MDS, between 2001–2004 (n = 581) and 2006–2011 (n = 1295) [18]. Although there was an increase in HMA use from 4% for the earlier period (before the approval of azacitidine and decitabine) to 43% for the later period, these results suggest that over half of higher-risk patients did not receive an HMA after approval of these agents. A second study, also based on data from the SEER-Medicare database for patients diagnosed with RAEB but covering a later period of January 2011 to December 2015 (n = 1190), similarly observed that 44% of patients did not receive HMAs [19]. These authors further reported that statistically significant risk factors for not receiving HMAs were: older age at diagnosis (e.g., 66–70 years vs ≥80 years; odds ratio [OR]: 2.36 [95% CI: 1.56–3.56]), being single (single vs married; OR: 0.67 [95% CI: 0.51–0.89]), having more co-morbidities (≥3 for National Cancer Institute co-morbidity score vs 0–1; OR: 0.62 [95% CI: 0.46–0.83]), and having a poor performance status (poor vs good; OR: 0.67 [95% CI: 0.51–0.87]). The influence of age and marital status is supportive of the hypothesis that logistical factors could contribute to poor utilization of HMA therapy. A more recent study analyzing data from the Optum® Research Database for patients newly diagnosed with higher-risk MDS between January 2008 and January 2016 (n = 209) observed that under two-thirds (62%) of patients received any MDS-related treatment, of whom 90% received HMAs in the first-line setting; consistent with other studies, this corresponds to 44% not receiving HMA therapy [25].
| Author (year) | Data source | Population | HMA use (%) | Definition of nonpersistence | Proportion of patients who were nonpersistent (%) | Ref. |
|---|---|---|---|---|---|---|
| Bell et al. (2019) | Optum database, 2008–2015 | Patients diagnosed with higher-risk MDS, n = 335 | MDS-related therapy received by 62.4% of patients; 89.5% of patients receiving MDS-related first-line therapy received HMAs | NR | NR | [25] |
| Cheng et al. (2021) | IBM MarketScan Commercial Claims and Encounters database and the Medicare Supplemental and Coordination of Benefits database, 2011–2018 | Diagnosed with MDS and starting HMA therapy, n = 2400 | NA (treated population) | Gap of ≥60 days before the end of the landmark period | 4-month landmark period, 18.8% 9-month landmark period, 43.7% | [36] |
| Cogle et al. (2017) | Optum Clinformatics Data Mart, 2009–2011 | Incident cases with MDS, n = 4151 | Initiating HMA therapy: 2.3% in year 1 2.7% in year 2 2.9% in year 3 | NR | NR | [30] |
| Corman et al. (2021) | SEER-Medicare database, January 2011–December 2015 | Diagnosed with RAEB, n = 1190 | 56.0% (ever prescribed) | Less than four cycles or a gap of ≥90 days between consecutive cycles | 44.6% | [19] |
| Davidoff et al. (2020) | SEER-Medicare database, 2001–2004 and 2006–2011 | Diagnosed with RAEB 2001–2004, n = 581; 2006–2011, n = 1295 | 2001–2004, 3.6% 2006–2011, 43.0% | NR | NR | [18] |
| Demakos et al. (2014) (abstract) | Claims, database not specified, 2009–2011 | Newly diagnosed with MDS | 13.1% | Less than six cycles | 69.1% | [26] |
| Ma et al. (2018) | GE Centricity Electronic Medical Record database, 2006–2014 | Patients with MDS, n = 5162 Patients who received ≥1 erythropoiesis-stimulating agent, iron chelation therapy, lenalidomide or HMA, n = 2079 | Among patients receiving ≥1 therapy, 12.1% received HMA first-line, 6.2% received HMA second- or third-line | NR | NR | [29] |
| Mukherjee et al. (2014) | US commercial claims database, 2009–2012 | Patients who received HMA, n = 1366 | NA (treated population) | Less than five cycles azacitidine Less than five cycles decitabine | 48.0% 52.0% | [35] |
| Sekeres et al. (2008) | Surveys of US hematologists and medical oncologists, June 2005–January 2007, n = 101 | Patients with MDS seen by participants: recently diagnosed, n = 670; established patients, n = 3844 | Newly diagnosed vs established patients: azacitidine, 16 vs 11–15%; decitabine, 2 vs 0–4% | NR | NR | [27] |
| Steensma et al. (2014) | Survey | Patients with MDS, n = 477; physicians managing MDS, n = 120 | 35% of patients | Less than six cycles azacitidine Less than four cycles decitabine | 41% 33% | [28] |
| Stein et al. (2019) | SEER-Medicare database, 2006–2017 | Diagnosed with MDS 2009–2017 and treated with HMAs, n = 3046 | NA (treated population) | Less than four cycles | 45.3% | [20] |
| Zeidan et al. (2020) | SEER-Medicare database | Diagnosed 2004–2013 and received HMA therapy, n = 2086 | NA (treated population) | Less than four cycles Less than six cycles | 42.7% 60.6% | [22] |
Other studies have reported the proportion of the overall MDS population (regardless of risk category) receiving HMA therapy, with rates ranging from 13–18% in newly diagnosed patients [26,27] to 11–19% [27] and 35% [28] in patients with an established diagnosis or where time from diagnosis was not stated. In another study, among patients treated with at least one of the following – erythropoiesis-stimulating agents, iron chelation therapy, lenalidomide or HMAs – only 18% received HMAs [29]. Given that more than 40% of patients have higher-risk disease [7], these data further suggest that less than half of patients in whom HMA therapy is recommended receive HMAs. Two of these studies were based on analysis of data from US claims databases [26,29], and two were based on US physician surveys [27,28]. A further study of incident cases of MDS in the USA between 2009 and 2011 reported that only 2% of patients initiated HMA therapy in the first year of diagnosis, with 3% initiating HMA therapy in each of years 2 and 3, suggesting that by the end of the third year, approximately 8% of patients had received HMA therapy; the reason for this lower rate compared with other studies is unclear [30].
Real-world persistence with HMA therapy in higher-risk MDS relative to treatment guidelines
In addition to underutilization of HMA therapy in patients with higher-risk MDS, a number of the studies described above and several further studies report low rates of persistence with HMA therapy (Table 1). Patients generally can require at least four to six cycles (in the absence of clear progression or unacceptable toxicity) in order to achieve responses to HMA therapy, as reflected in the prescribing information for azacitidine and decitabine, along with guidelines from the National Comprehensive Cancer Network (NCCN) and the European Society for Medical Oncology, and further improvements in outcomes may be achieved beyond this time point [12,13,23,24,31]. For example, in the AZA-001 trial, although 91% of patients achieved a response after six cycles, in 39% of patients this was not their best response, whereas by cycle 12, 92% of patients had achieved their best response. The median number of cycles among responders was 14 [32]. Furthermore, the HRQoL benefits with azacitidine treatment were greater in patients who received at least four cycles of therapy [33]. Among patients with higher-risk MDS treated with three different dose schedules of decitabine, an overall response rate of 70% (complete response rate: 35%) was achieved with a median of ≥7 cycles, with patients with ≥6 months' follow-up receiving a median of ≥10 cycles [34].
Real-world studies indicate that a considerable proportion of patients are not persistent with HMA therapy, when defined as receiving less than four or less than six cycles, or having a gap of 90 days between cycles. Four studies have reported that around 33–45% of patients received less than four cycles of therapy [19,20,22,28], and four studies reported that 41–69% received less than five or six cycles [22,26,28,35]. Data for patients in the SEER-Medicare database diagnosed with MDS (2009–2015) and treated with HMAs (n = 3046) reported a mean of 5.1 cycles (median: 4) [20]. In another study based on data from the IBM® MarketScan® Commercial Claims and Encounters database and the Medicare Supplemental and Coordination of Benefits database (2011–2018), among 1929 patients with MDS starting HMA therapy, 19% were considered nonpersistent (defined by the start of a gap of ≥60 days between HMA therapy claims) within 2 months of HMA initiation [36]. Median time to treatment discontinuation (i.e., persistence) in this study was 5.6 months [36].
Impact of persistence with HMA therapy on clinical outcomes
Based on the clinical trial evidence suggesting that patients require at least four to six cycles of therapy to achieve responses to HMA, nonpersistence in the real-world setting would be expected to impact clinical outcomes. This is evident from the results of four studies reporting outcomes according to the duration of therapy (Table 2).
| Author (year) | Data source | Population | Definition of persistence/ nonpersistence | Clinical outcomes, nonpersistent vs persistent | Ref. |
|---|---|---|---|---|---|
| Cabrero et al. (2015) | Clinical trials of HMA | Patients who stopped therapy while in response, n = 16 | <12 vs 12 cycles | 1-year PFS: 17 vs 50%, p = 0.062 Median OS: 4 vs 20 months, p = 0.043 | [39] |
| Cheng et al. (2021) | IBM MarketScan Commercial Claims and Encounters database and the Medicare Supplemental and Coordination of Benefits database from 2011 to 2018 | Diagnosed with MDS and starting HMA therapy, n = 2400 | Nonpersistence, gap of ≥60 days in treatment before end of landmark period | Mean time to AML transformation, 22.0 vs 38.5 months Incidence rate of AML during follow-up from HMA initiation, adjusted HR: 1.88; 95% CI: 1.53–2.32; p < 0.001 | [36] |
| Corman et al. (2021) | SEER-Medicare database | Diagnosed with RAEB between January 2011 and December 2015, n = 1190 | Less than four cycles or a gap of ≥90 days between consecutive cycles | Median OS, 9.5 vs 13.8 months (No HMA therapy, 3.8 months) | [19] |
| Zeidan et al. (2020) | SEER-Medicare database | Diagnosed 2004–2013 and received HMA therapy, n = 2086 | Less than four vs ≥4 cycles | Median OS, 4 vs 16 months; p < 0.01 | [22] |
A claims analysis of MDS patients diagnosed between 2004 and 2013 (SEER-Medicare database; n = 2086) and who received HMA therapy found that 61% of patients received less than six cycles and 43% received less than four cycles of HMA therapy [22]. Median OS was significantly shorter in patients who received less than four compared with four or more cycles (4 vs 16 months; p < 0.01). Similarly, a claims analysis based on data from the SEER-Medicare database for patients diagnosed with RAEB between January 2011 and December 2015 (n = 1190) estimated that 44% of patients were never prescribed HMA therapy, while 25% were nonpersistent (defined as having received less than four cycles or had a gap of ≥90 days between consecutive cycles) and only 31% were persistent with therapy [19]. In this study, median OS was also longer in persistent versus nonpersistent patients (13.8 vs 9.5 months) and, notably, was shortest for patients who had never received HMA therapy (3.8 months) [19].
A retrospective study using the IBM MarketScan Commercial Claims and Encounters database and the Medicare Supplemental and Coordination of Benefits database from 2011 to 2018 (n = 2400 patients with MDS starting HMA therapy) found that the incidence of transformation from MDS to AML was almost twofold higher in nonpersistent versus persistent patients (hazard ratio [HR]: 1.88; 95% CI: 1.53–2.32; p < 0.001) during follow-up from the time of HMA persistence, where nonpersistence was defined as having a gap of ≥60 days in treatment before the end of a landmark period of 4 months [36]. Furthermore, the mean time to AML transformation was shorter in nonpersistent patients (22.0 vs 38.5 months) [36].
Another issue that may affect the real-world effectiveness of treatment is whether therapy is delayed, thus extending the duration of a treatment cycle. This may occur for clinical reasons such as management of toxicities, but it may also occur because of logistical difficulties associated with the need for administration in the outpatient setting. This has been highlighted by a single-center study of azacitidine, which found that 31% of cycles were delayed by more than 28 days, with 8% of cycles being delayed more than 42 days [37].
The clinical importance of persistence with HMA therapy has also been demonstrated in various studies, showing that discontinuation of therapy in responding patients results in rapid disease progression and a poor prognosis. Voso et al. [38] observed rapid disease progression in 13 patients who discontinued azacitidine while still responding to treatment. Median time to progression was 5.4 months and median OS from discontinuation was 6.6 months. Reasons for discontinuation were patient choice (n = 3), co-morbid conditions (n = 6) and adverse events such as cutaneous reactions (n = 2). An analysis of data for 16 patients included in early trials of HMAs also showed that patients relapse rapidly after discontinuing therapy [39]. Median progression-free survival (PFS) was 4 months and median OS from discontinuation was 16 months. Furthermore, median OS was significantly shorter in patients who stopped therapy while still in response after less than 12 versus ≥12 cycles (4 vs 20 months; p = 0.043), and 1-year PFS was also lower in those who stopped therapy after less than 12 months (1-year PFS: 17 vs 50%; p = 0.062). Rapid disease progression on discontinuation of therapy reflects the mechanism of action of HMAs, the effects of which, unlike cytotoxic agents, are rapidly reversible, thus necessitating continuous administration for continued efficacy [40].
Factors influencing persistence with HMA therapy
Early discontinuation of HMA therapy can occur for reasons that are clinically driven (e.g., toxicity or disease progression) or nonclinically driven (e.g., factors relating to the logistics of therapy administration, provider inexperience or social/economic determinants of health) [19,41]. Although improving the management of adverse events could offer some improvement in HMA persistence [42], addressing nonclinical reasons for discontinuation could further improve real-world outcomes and therefore should also be a key consideration for clinicians and healthcare decision-makers.
The need for frequent hospital visits to receive therapy that has to be administered by a healthcare professional, and especially the need for lengthy infusions, can be one reason for poor persistence. Frequent hospital visits can be particularly difficult for elderly patients, some of whom may be unable to travel to hospital alone and may not have local family members or friends who can assist with hospital visits. Indeed, in elderly patients with lung cancer, the need for hospitalization has been shown to be a factor in the decision to decline chemotherapy [43]. Therapies requiring hospital visits present logistical/administrative challenges for clinics, and can impact families/caregivers, also leading to poor persistence. Administrative challenges may be heightened when therapy should be given on 5 or 7 consecutive days, where public holidays may often interfere with scheduling. These challenges have been highlighted by an Italian study, which found that of 56 cycles of azacitidine for which administration was delayed, 27% were delayed because of organizational problems at centers (public holidays), and 11% because of personal and family difficulties, largely related to taking time off work for appointments [37].
In addition to the logistical difficulties associated with the need for hospital attendance, IV and SC routes of administration are associated with inherent challenges, such as the psychological distress and potential medical complications associated with venous access, and for SC therapies, injection-site reactions and pain. These too may affect patient quality of life and potentially impact persistence [44,45].
Other reasons for poor persistence may relate to physician perceptions of the burden of treatment and appreciation of the need for prolonged therapy to achieve optimum benefit and sustain responses. A survey of US physicians and their patients with MDS found that 79% of physicians reported recommending early discontinuation of therapy due to adverse events, while 69% did so because they considered the burden of therapy outweighed the benefits [28]. However, the corresponding rates for patients were 33 and 35%, respectively, suggesting that patients were more prepared to tolerate the burden of treatment.
The low incidence of higher-risk MDS also means that most physicians treat only a limited number of patients with this malignancy. Given the distinct mechanism of action of HMAs that necessitates continuous and prolonged therapy for patients to achieve an optimal response, persistence may be influenced by the treating physician’s limited experience with HMAs and lack of awareness of the need for prolonged continuous therapy or how best to manage adverse events. For example, Zeidan et al. [22] investigated the possible impact of physician experience with HMAs on persistence using data from the SEER-Medicare database for patients diagnosed between 2004 and 2013 (n = 2086). They found that patients who were managed by physicians who had started at least one patient on HMA therapy in the last year were significantly more likely to receive four or more cycles compared with those patients managed by physicians who had less experience with HMAs (OR: 1.29; 95% CI: 1.06–1.57; p = 0.01). In addition, patients who had received three or more platelet transfusions were less likely to be persistent than those who had received no platelet transfusions (OR: 0.53; p = 0.001), as were patients who had received three or more RBC transfusions versus no RBC transfusions (OR: 0.74; p = 0.016). This is also consistent with findings from a subgroup analysis of a Phase III trial comparing guadecitabine and physician’s choice as treatment for AML, which reported a trend toward better outcomes for patients treated at centers that had participated in earlier studies with this HMA [46].
Economic consequences of nonpersistence with HMA therapy
In addition to the clinical impact of nonpersistence with HMA therapy, persistence has also been found to have an economic impact as observed in three studies comparing resource use in persistent versus nonpersistent patients (Table 3).
| Author (year) | Data source | Population | Definition of persistence/ nonpersistence | Economic outcomes | Ref. |
|---|---|---|---|---|---|
| Cheng et al. (2021) | IBM MarketScan Commercial Claims and Encounters database and the Medicare Supplemental and Coordination of Benefits database, 2011–2018 | Patients with MDS starting HMA therapy, n = 2400 | ≥60-day gap in treatment before the end of 4-month landmark period | Nonpersistent vs persistent: Significantly higher all-cause HRU for: ER visits (IRR: 1.16; 95% CI: 1.01–1.34), inpatient visits (IRR: 1.46; 95% CI: 1.28–1.67); and inpatient days (IRR: 1.40; 95% CI: 1.33–1.46); all p < 0.001 Significantly higher non-HMA-related HRU burden for: ER visits (IRR: 1.30; 95% CI: 1.12–1.50), inpatient visits (IRR: 1.48; 95% CI: 1.30–1.69), inpatient days (IRR: 1.41; 95% CI: 1.36–1.46) and outpatient visits (IRR: 1.12; 95% CI: 1.10–1.14); all p < 0.001 Fewer HMA-related outpatient visits (IRR: 0.09; 95% CI: 0.09–0.10) and marginally fewer any-cause outpatient visits (IRR: 0.82; 95% CI: 0.80–0.83); all p < 0.001 | [36] |
| Cogle et al. (2017) | Optum Clinformatics Data Mart, 2008–2009 | Patients with refractory MDS following HMA therapy, n = 402 | Stopped HMA therapy | Total healthcare costs following HMA failure: US$76,945 (SD US$92,764) during the first 6 months (n = 402); US$50,732 (SD US$77,885) for months 19–24 (n = 95) | [30] |
| Joshi et al. (2021) | SEER-Medicare database, 2011–2016 | Patients diagnosed with RAEB and who received HMAs, n = 664 | Less than four cycles or a gap of ≥90 days between cycles | Nonpersistent vs persistent: Significantly higher hospitalizations (IRR: 1.54; p = 0.001), ER visits (IRR: 1.32; p < 0.001), skilled nursing facility use (IRR: 2.16; p = 0.003), home health visits (IRR: 1.34; p = 0.024) and hospice care use (IRR: 2.56; p < 0.001). Significantly lower frequency of outpatient (IRR: 0.87; p = 0.026) and physician visits (IRR: 1.23; p < 0.001) Significantly (p < 0.05) higher total PPPM costs (US$18,039 vs US$13,893), particularly for hospitalizations (US$3375 vs US$2131), and ER costs (US$5517 vs US$2867) | [47] |
| Stein et al. (2021) | SEER-Medicare database, 2006–2016 | Patients diagnosed with MDS and initiated on HMA therapy, n = 3046 | Treatment success, defined as receipt of ≥7 cycles, SCT or RBC transfusion independence Treatment failure, defined as disease progression, HMA discontinuation, resumption of RBC transfusion dependence, AML or death | Treatment success vs treatment failure (pre-HMA) vs treatment failure (post-HMA), per 100 patients per month: Inpatient admissions, 7.5 vs 20.4 vs 35.3 Total healthcare costs: US$8069 vs US$13,809 vs US$19,242 Outpatient costs: US$7028 vs US$9099 vs US$3702 Inpatient costs: US$1002 vs US$4616 vs US$15,451 | [48] |
A retrospective cohort study using data from the IBM MarketScan Commercial Claims and Encounters database and the Medicare Supplemental and Coordination of Benefits database from 2011 to 2018 (n = 2400 with MDS starting HMA therapy) compared resource use and costs among patients who were persistent versus non-persistent with HMA therapy at 4 months (n = 1534 vs 363) [36]. Nonpersistent patients relative to persistent patients had a significantly higher (p < 0.001) non-HMA-related healthcare resource utilization burden, particularly for ER visits, inpatient visits and inpatient days, although, as expected, nonpersistent patients had fewer HMA-related outpatient visits and marginally fewer any-cause outpatient visits. Use of blood transfusions and growth factors was similar between the two groups [36].
A second retrospective cohort study identified patients from the SEER-Medicare-linked database diagnosed with RAEB between 2011 and 2016 who received HMAs, and found per-patient-per-month (PPPM) resource utilization was significantly higher among HMA nonpersistent patients (defined as having received less than four cycles or a gap of ≥90 days between cycles) compared with HMA-persistent patients, for hospitalizations (incidence rate ratio [IRR]: 1.54; p = 0.001), ER visits (IRR: 1.32; p < 0.001), skilled nursing facility use (IRR: 2.16; p = 0.003), home health visits (IRR: 1.34; p = 0.024) and hospice care use (IRR: 2.56; p < 0.001). In contrast, as expected, the frequency of outpatient (IRR: 0.87; p = 0.026) and physician visits (IRR: 1.23; p < 0.001) was significantly higher among HMA-persistent patients than nonpersistent patients. The HMA nonpersistent group (n = 295) incurred significantly (p < 0.05) higher total PPPM costs compared with the HMA persistent group (n = 369; US$18,039 vs US$13,893), particularly for hospitalizations (US$3375 vs US$2131) and ER costs (US$5517 vs US$2867) [47].
The third study compared resource use and costs for patients experiencing treatment success (defined as receipt of ≥7 cycles, stem cell transplantation or RBC transfusion independence) or treatment failure (defined as disease progression, HMA discontinuation, resumption of RBC transfusion dependence, AML or death) in patients diagnosed with MDS and initiated on HMA therapy (n = 3046) [48]. Based on data from the SEER-Medicare database (2006–2016), the authors found that hospital admissions were considerably lower for patients with successful HMA therapy (7.5 per 100 patients per month) compared with those with treatment failure both pre- and post-HMA therapy (20.4 and 35.3 per 100 patients per month, respectively). Total healthcare costs showed a similar trend, being over twofold higher for patients with treatment failure (post-HMA therapy) compared with patients with treatment success (US$19,242 vs US$8069 PPPM). The difference in costs largely reflected 15-fold higher inpatient costs for the treatment failure group [48].
Another study has investigated the resource use and costs for management of patients who discontinued initial HMA therapy [30]. The authors analyzed data from the Optum Clinformatics® Data Mart (2008–2009) for 402 patients with refractory MDS following HMA therapy. Total healthcare costs after HMA failure were high (US$76,945) during the first 6 months after failure and were primarily driven by nonpharmacy costs, which accounted for approximately 90% of the total healthcare costs. Total costs in the following 6-month periods remained greater than US$50,000, despite the number of patients decreasing progressively in each time period.
Improving real-world clinical & economic outcomes with HMA therapy
The evidence reviewed suggests that underutilization and poor persistence are likely to be important factors contributing to the suboptimal outcomes for HMA therapy seen in patients with MDS in routine clinical practice, with persistence impacting survival outcomes and HRQoL, and contributing to increased healthcare resource use and costs. These data indicate the need to understand the drivers of underutilization and poor persistence with HMA therapy, and for approaches to improve utilization and persistence, and to ensure that patients achieve the optimum benefit from HMA therapy. It should be noted that HMA therapy is not a curative therapy and effectiveness is limited for a proportion of patients for various reasons, including the heterogeneity of the disease. Although there remains a need for more effective therapies in the treatment of MDS, in their absence it is important to maximize the effectiveness of currently available options.
One such approach may be the use of oral HMA therapies. Recently, an oral formulation of decitabine – oral decitabine and cedazuridine (formerly referred to as ASTX727) – has been approved by the FDA for patients with higher-risk MDS, and is included in NCCN guidelines as an alternative to IV and SC HMAs [13,49]. Oral decitabine and cedazuridine has demonstrated systemic exposure equivalent to IV decitabine with a similar response rate and adverse event profile in Phase II and III studies [49–51]. An oral formulation of azacitidine was also recently approved by the FDA, although it is not pharmacokinetically equivalent to the IV or SC formulation and is not indicated for the treatment of MDS [52]. Such oral formulations may overcome some of the logistical challenges hypothesized to contribute to the suboptimal outcomes of HMA therapy in routine practice. This is supported by a body of published literature, as reviewed by Eek et al. [53], indicating that oncology patients largely prefer oral versus IV therapy because of the ability to take medication at home and avoid the need for IV access or injections. While further research is needed on the impact of oral versus IV/SC therapy on persistence and adherence with HMA therapy, evidence suggests that adherence to oral oncology drugs is generally good [54–56], and one study in metastatic renal cell carcinoma reported higher persistence with oral therapy than IV therapy [57]. A study in non-small-cell lung cancer has also shown oral therapy administered at home to be associated with healthcare cost savings compared with IV therapy [58].
An oral therapy may further enable physicians to offer HMA therapy to patients for whom the administration of IV or SC regimens may be considered to place too great a burden on patients and their families. This is particularly relevant in the current environment where there is a move to favor treatments that can be administered at home and avoid the need for hospital visits and hospitalization during the coronavirus pandemic, as supported by guidance from both the FDA and the NCCN Best Practices Committee [59,60].
Other approaches that may help improve persistence with HMAs involve physician education on the need for prolonged and continuous treatment with HMAs, how best to manage toxicities, and ensuring physicians have an appreciation of patient attitudes regarding the burden of treatment versus the therapeutic benefits of active treatment. Patients and their caregivers also need to understand the need for prolonged continuous therapy to ensure that they make efforts to adhere to their therapy insofar as this is clinically possible.
Limitations
The evidence discussed in this review provides a largely consistent picture of underuse and poor persistence with HMAs for the management of higher-risk MDS, and the impact of poor persistence on clinical and economic outcomes. However, the findings should be interpreted with consideration to the limitations of this research. First, most of the identified studies of persistence with HMAs and their impact on clinical and economic outcomes were performed in the USA and therefore may not be generalizable to other countries. Second, some of the identified studies on persistence and the clinical impact of poor persistence are based on data collected over 10 years ago and so may not be fully representative of current treatment practice. However, it is recognized that few new therapies have been introduced over this period, suggesting that substantial changes in clinical practice are unlikely to have occurred. Third, the need to treat with HMAs to disease progression has become better understood over time and would be expected to result in a trend toward better persistence; however, this is not evident in the available literature. Fourth, risk classification for MDS has changed over the last 2 decades and may have influenced the proportion of patients considered to be eligible for HMA therapy. Finally, this review is based on evidence identified through targeted searches, and therefore may not have identified all available evidence for persistence with HMA therapy in patients with MDS and its impact on clinical and economic outcomes. However, the findings from the identified studies, as described here, are generally consistent with each other, suggesting they are likely to be representative of the available evidence on this topic.
Conclusion & future perspective
Improving the outlook for the growing number of older adults being diagnosed with higher-risk MDS is an important treatment goal that may require new approaches. Clinical trial data for the HMAs azacitidine and decitabine have demonstrated clinically meaningful improvements with these agents over supportive care; however, the outcomes reported for real-world studies fall short of those achieved in clinical trials. The introduction of oral HMA therapy, as an alternative to IV or SC treatments, offers a potential option to help address some of the challenges with persistence to HMA therapy in clinical practice, as well as to avoid the risks and burden associated with in-person clinic visits by allowing administration at home. Further studies are needed to understand the potential real-world benefit of oral HMA therapy on persistence and, relatedly, on clinical and economic outcomes.
Real-world utilization & persistence of hypomethylating agent therapy for higher-risk myelodysplastic syndromes
Data from real-world studies suggest that around half of patients with higher-risk myelodysplastic syndromes (MDS) do not receive hypomethylating agent (HMA) therapy.
In addition to underutilization, real-world studies report low rates of persistence with HMA therapy.
Patients generally require at least four to six cycles in order to achieve responses to HMA therapy, and further improvements in outcomes may be achieved beyond this time point; however, studies have reported that around 33–45% of patients received less than four cycles of therapy, and that 41–69% received less than five or six cycles.
Impact of persistence with HMA therapy on clinical outcomes
In real-world studies, nonpersistence with HMA therapy has been associated with more rapid disease progression, shorter overall survival and higher risk of progression from MDS to acute myeloid leukemia.
Factors influencing persistence with HMA therapy
Early discontinuation of HMA therapy can occur for reasons that may be considered clinically driven (e.g., toxicity or disease progression) or nonclinically driven (e.g., factors relating to the logistics of therapy administration, or provider inexperience).
Alongside clinical factors, addressing nonclinical reasons for discontinuation could further improve real-world outcomes, and therefore should also be a key consideration for clinicians and healthcare decision-makers.
Economic consequences of nonpersistence with HMA therapy
Persistence with HMA therapy among patients with MDS has also been found to have an economic impact, with nonpersistence associated with higher healthcare resource utilization and costs.
Improving real-world clinical & economic outcomes with HMA therapy
The data reviewed indicate the need to understand the drivers of underutilization and poor persistence with HMA therapy, and for approaches to improve utilization and persistence in real-world clinical practice to ensure patients achieve the optimum benefit from HMA therapy.
Potential approaches include the use of oral HMA therapies, and improving clinician, patient and caregiver understanding of the need for prolonged use of HMA therapy.
Financial & competing interests disclosure
Taiho Oncology funded the targeted literature review that was used as a basis for this article and reviewed the draft manuscript for medical accuracy. AM Zeidan is a Leukemia and Lymphoma Society Scholar in Clinical Research; has received research funding (institutional) from AbbVie, ADC Therapeutics, Amgen, Aprea, Astex, Boehringer Ingelheim, Celgene/BMS, Incyte, MedImmune/AstraZeneca, Novartis, Pfizer, Takeda and Trovagene/Cardiff Oncology; has participated in advisory boards, and/or had a consultancy with and received honoraria from AbbVie, Acceleron, Agios, Amgen, Aprea, Astellas, BeyondSpring, Boehringer Ingelheim, Cardinal Health, Celgene/BMS, Daiichi Sankyo, Epizyme, Genentech, Gilead, Incyte, Ionis, Janssen, Jazz, Kura, Loxo Oncology, Novartis, Otsuka, Pfizer, Seagen, Syndax, Taiho, Takeda, Trovagene/Cardiff Oncology and Tyme; has served on clinical trial committees for AbbVie, Celgene/BMS, Geron, Gilead, Kura, Loxo Oncology and Novartis; and has received travel support for meetings from Cardiff Oncology, Novartis and Pfizer. T Salimi is an employee of Taiho Oncology. R Epstein is an employee of Epstein Health LLC, paid consultants to Taiho Oncology in connection with this study, and holds leadership positions in Decipher Biosciences, Fate Therapeutics, Illumina, Proteus Digital Health and Veracyte; and has received consulting fees from Halozyme, Intra-Cellular Therapies, Merck, Otsuka, Radius Health and Taiho Oncology. 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.
Medical writing support was provided by Rowena Hughes and Elizabeth Harvey of Curo (a division of Envision Pharma Group), and funded by Taiho Oncology.
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/
Papers of special note have been highlighted as: • of interest; •• of considerable interest
References
- 1. . Chapter e154. Myelodysplastic syndromes. In: Pharmacotherapy: A Pathophysiologic Approach (11th Edition). Dipiro JT (Ed.) McGraw-Hill, NY, USA (2020).
- 2. . Myelodysplastic syndromes current treatment algorithm 2018. Blood Cancer J. 8(5), 47 (2018).
- 3. . Health care utilization and risk of infection and bleeding among patients with myelodysplastic syndromes with/without transfusions, and with/without active therapy. Leuk. Lymphoma 55(5), 1119–1125 (2014).
- 4. Surveillance Epidemiology and End Results Program. SEER*Explorer. Myelodysplastic syndromes (MDS) (2020). https://seer.cancer.gov/explorer/index.html
- 5. . Epidemiology of myelodysplastic syndromes: why characterizing the beast is a prerequisite to taming it. Blood Rev. 34, 1–15 (2019).
- 6. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 89(6), 2079–2088 (1997).
- 7. Revised international prognostic scoring system for myelodysplastic syndromes. Blood 120(12), 2454–2565 (2012).
- 8. Impact of the degree of anemia on the outcome of patients with myelodysplastic syndrome and its integration into the WHO classification-based Prognostic Scoring System (WPSS). Haematologica 96(10), 1433–1440 (2011).
- 9. Time-dependent prognostic scoring system for predicting survival and leukemic evolution in myelodysplastic syndromes. J. Clin. Oncol. 25(23), 3503–3510 (2007).
- 10. . Myelodysplastic syndromes: 2018 update on diagnosis, risk-stratification and management. Am. J. Hematol. 93(1), 129–147 (2018).
- 11. Causes of death in 2877 patients with myelodysplastic syndromes. Ann. Hematol. 95(6), 937–944 (2016).
- 12. Myelodysplastic syndromes: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up †. Ann. Oncol. 32(2), 142–156 (2021).
- 13. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): myelodysplastic syndromes Version 3.2021 (2021). www.nccn.org/professionals/physician_gls/pdf/mds.pdf
- 14. Health Canada. Drug product database (2021). www.canada.ca/en/health-canada/services/drugs-health-products/drug-products/drug-product-database.html
- 15. Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J. Clin. Oncol. 20(10), 2429–2440 (2002).
- 16. Decitabine improves patient outcomes in myelodysplastic syndromes: results of a Phase III randomized study. Cancer 106(8), 1794–1803 (2006).
- 17. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, Phase III study. Lancet Oncol. 10(3), 223–232 (2009).
- 18. Hypomethylating agent (HMA) therapy use and survival in older adults with Refractory Anemia with Excess Blasts (RAEB) in the United States (USA): a large propensity score-matched population-based study †. Leuk. Lymphoma 61(5), 1178–1187 (2020).
- 19. . Under-use of hypomethylating agents in patients with higher-risk myelodysplastic syndrome in the United States: a large population-based analysis. Clin. Lymphoma Myeloma Leuk. 21(2), e206–e211 (2021). •• Assessed utilization and persistence with hypomethylating agent (HMA) therapy among patients with higher-risk myelodysplastic syndrome (MDS) in the USA.
- 20. Treatment patterns and outcomes in patients with myelodysplastic syndromes treated with hypomethylating agents: a SEER-Medicare analysis. Blood 134(Suppl. 1), 3495 (2019). • Assessed persistence with HMA therapy among patients with MDS in the USA.
- 21. Modest improvement in survival of patients with refractory anemia with excess blasts in the hypomethylating agents era in the United States. Leuk. Lymphoma 58(4), 982–985 (2017).
- 22. Association of provider experience and clinical outcomes in patients with myelodysplastic syndromes receiving hypomethylating agents. Leuk. Lymphoma 61(2), 397–408 (2020). • Assessed the association between provider experience and persistence with HMA therapy and survival among patients with MDS.
- 23. Celgene Corporation. Azacitidine (Vidaza) prescribing information (2020). https://packageinserts.bms.com/pi/pi_vidaza.pdf
- 24. Otsuka. Decitabine (Dacogen) (2018). www.accessdata.fda.gov/drugsatfda_docs/label/2018/021790s021lbl.pdf
- 25. Economic burden of patients treated for higher-risk myelodysplastic syndromes (HR-MDS) in routine clinical care in the United States. Pharmacoecon. Open 3(2), 237–245 (2019).
- 26. Incidence and treatment of myelodysplastic syndrome in the US: treatment approaches, optimization of care and the need for additional therapeutic agents. Blood 124(21), 1287 (2014).
- 27. Characteristics of US patients with myelodysplastic syndromes: results of six cross-sectional physician surveys. J. Natl Cancer Inst. 100(21), 1542–1551 (2008).
- 28. Disparity in perceptions of disease characteristics, treatment effectiveness, and factors influencing treatment adherence between physicians and patients with myelodysplastic syndromes. Cancer 120(11), 1670–1676 (2014).
- 29. . Selection of patients with myelodysplastic syndromes from a large electronic medical records database and a study of the use of disease-modifying therapy in the United States. BMJ Open 8(7), e019955 (2018).
- 30. The incidence and health care resource burden of the myelodysplastic syndromes in patients in whom first-line hypomethylating agents fail. Oncologist 22(4), 379–385 (2017).
- 31. Outcomes of patients with myelodysplastic syndromes who achieve stable disease after treatment with hypomethylating agents. Leuk. Res. 41, 43–47 (2016).
- 32. Continued azacitidine therapy beyond time of first response improves quality of response in patients with higher-risk myelodysplastic syndromes. Cancer 117(12), 2697–2702 (2011).
- 33. Impact of azacytidine on the quality of life of patients with myelodysplastic syndrome treated in a randomized Phase III trial: a Cancer and Leukemia Group B study. J. Clin. Oncol. 20(10), 2441–2452 (2002).
- 34. Update of the decitabine experience in higher risk myelodysplastic syndrome and analysis of prognostic factors associated with outcome. Cancer 109(2), 265–273 (2007).
- 35. Treatment patterns among patients with myelodysplastic syndromes: observations of 1st-line therapy, discontinuation and the need of additional therapies. Blood 124(21), 2598–2598 (2014).
- 36. Persistence to hypomethylating agents and clinical and economic outcomes among patients with myelodysplastic syndromes. Hematology (Amsterdam, Netherlands) 26(1), 261–270 (2021). •• Evaluated the association of HMA persistence with healthcare resource use and clinical outcomes among patients with MDS.
- 37. Obstacles to adherence to azacitidine administration schedule in outpatient myelodysplastic syndrome and related disorders. Support. Care Cancer 23(2), 303–305 (2015).
- 38. Rapid loss of response after withdrawal of treatment with azacitidine: a case series in patients with higher-risk myelodysplastic syndromes or chronic myelomonocytic leukemia. Eur. J. Haematol. 90(4), 345–348 (2013).
- 39. Discontinuation of hypomethylating agent therapy in patients with myelodysplastic syndromes or acute myelogenous leukemia in complete remission or partial response: retrospective analysis of survival after long-term follow-up. Leuk. Res. 39(5), 520–524 (2015).
- 40. . How we treat higher-risk myelodysplastic syndromes. Blood 123(6), 829–836 (2014).
- 41. The impact of social determinants of health on medication adherence: a systematic review and meta-analysis. J. Gen. Intern. Med. 36(5), 1359–1370 (2021).
- 42. Management and supportive care measures for adverse events in patients with myelodysplastic syndromes treated with azacitidine*. Eur. J. Haematol. 85(2), 130–138 (2010).
- 43. . Incidence of chemotherapy discontinuation and characteristics of elderly patients with non-small cell lung cancer treated with platinum-based doublets. Contemp. Oncol. (Pozn.) 18(5), 340–343 (2014).
- 44. . Answering patients’ needs: oral alternatives to intravenous therapy. Oncologist 6(Suppl. 4), 12–16 (2001).
- 45. . Subcutaneous administration of anticancer agents. Anticancer Res. 34(4), 1579–1586 (2014).
- 46. Results from a global randomized Phase III study of guadecitabine (G) vs treatment choice (TC) in 815 patients with treatment naïve (TN) AML unfit for intensive chemotherapy (IC) ASTRAL-1 study: analysis by number of cycles. Blood 134(Suppl. 1), 2591–2591 (2019).
- 47. . Direct medical costs associated with treatment nonpersistence in patients with higher-risk myelodysplastic syndromes receiving hypomethylating agents: a large retrospective cohort analysis. Clin. Lymphoma Myeloma Leuk. 21(3), e248–e254 (2021). • Assessed healthcare resource use and medical costs associated with early discontinuation of HMA therapy among higher-risk MDS patients.
- 48. Healthcare resource utilization and costs in patients with myelodysplastic syndromes treated with hypomethylating agents: a SEER-Medicare analysis. J. Med. Econ. 24(1), 234–243 (2021).
- 49. INQOVI® USPI. INQOVI® (decitabine and cedazuridine) tablets, for oral use. US highlights of prescribing information. Otsuka Pharmaceutical Co. Ltd, Taiho Oncology Inc, Japan, PrincetonNJ, USA (2020). www.inqovi.com/pi
- 50. Oral cedazuridine/decitabine for MDS and CMML: a Phase II pharmacokinetic/pharmacodynamic randomized crossover study. Blood 136(6), 674–683 (2020).
- 51. Pharmacokinetic exposure equivalence and preliminary efficacy and safety from a randomized cross over Phase III study (ASCERTAIN) of an oral hypomethylating agent ASTX727 (cedazuridine/decitabine) compared to IV decitabine (Oral). Presented at: 61st American Society of Hematology Annual Meeting. December 7-9, 2019, Orlando, FL, USA (2019).
- 52. Celgene Corporation. Onureg (oral azacitadine) prescribing information (2020). www.accessdata.fda.gov/drugsatfda_docs/label/2020/214120s000lbl.pdf
- 53. Patient-reported preferences for oral versus intravenous administration for the treatment of cancer: a review of the literature. Patient Prefer. Adherence 10, 1609–1621 (2016).
- 54. Treatment satisfaction and adherence to oral chemotherapy in patients with cancer. J. Oncol. Pract. 13(5), e474–e485 (2017).
- 55. Adherence and patients’ experiences with the use of oral anticancer agents. Acta Oncol. 53(2), 259–267 (2014).
- 56. Adherence to capecitabine treatment and contributing factors among cancer patients in Malaysia. Asian Pac. J. Cancer Prev. 15(21), 9225–9232 (2014).
- 57. Patterns of care in patients with metastatic renal cell carcinoma among a U.S. payer population with commercial or Medicare Advantage membership. J. Manag. Care Spec. Pharm. 22(3), 219–226 (2016).
- 58. Comparative cost-minimisation of oral and intravenous chemotherapy for first-line treatment of non-small cell lung cancer in the UK NHS system. Eur. J. Health Econ. 8(2), 145–151 (2007).
- 59. US Food and Drug Administration. FDA approves new therapy for myelodysplastic syndromes (MDS) that can be taken at home (2020). www.fda.gov/news-events/press-announcements/fda-approves-new-therapy-myelodysplastic-syndromes-mds-can-be-taken-home
- 60. Safety at the time of the COVID-19 pandemic: how to keep our oncology patients and healthcare workers safe. J. Natl Compr. Canc. Netw. 18(5), 504–509 (2020).
