Analysis of Real-World Data on Overall Survival in Multiple Myeloma Patients With ≥3 Prior Lines of Therapy Including a Proteasome Inhibitor (PI) and an Immunomodulatory Drug (IMiD), or Double Refractory to a PI and an IMiD

This retrospective study evaluated the treatment patterns in and overall survival of multiple myeloma patients who were refractory to a proteasome inhibitor and an immunomodulatory drug or who had received three or more prior lines of therapy. The findings of this survival analysis suggest that outcomes for these patients remain poor despite the availability of newer agents.

Abstract

Background.

This retrospective study evaluated the treatment patterns in and overall survival (OS) of multiple myeloma (MM) patients who were refractory to a proteasome inhibitor (PI) and an immunomodulatory drug (IMiD) or who had received three or more prior lines of therapy (LOTs) including a PI and an IMiD.

Methods.

Electronic health records in the IMS LifeLink and OPTUM databases were screened for indexing periods of 2000–2014 and 2007–2014, respectively. Patients who were refractory to both a PI and an IMiD (criterion 1) or who received three or more prior LOTs (including a PI and an IMiD) and showed disease progression within 60 days of their most recent regimen (criterion 2) comprised the eligible population. Median OS from time of last LOT was assessed for the full cohort, cohorts meeting criteria 1 and 2, and clinically important subgroups.

Results.

Of 3,929 and 3,837 patients with MM diagnoses evaluated in the IMS LifeLink and OPTUM databases, 500 and 162 met the eligibility criteria, respectively. Similar median OS was observed for eligible patients in the IMS LifeLink and OPTUM databases (7.9 vs. 7.9 months; p = .5358). In subgroup analyses of the IMS LifeLink data set, median OS was longer in patients <65 years of age than it was for those ≥65 years at eligibility (9.5 vs 6.7 months; p < .01) and in patients with good or unreported versus poor performance status at last claim (7.8 or 8.8 vs. 2.9 months; p < .0001).

Conclusion.

The findings of this survival analysis suggest that outcomes for these patients remain poor despite the availability of newer agents.

Implications for Practice:

This real-world retrospective study of electronic health records examines the survival outcomes of patients with multiple myeloma who are heavily pretreated or highly refractory to currently approved treatments, including recently approved proteasome inhibitors and immunomodulatory drugs. This survival analysis showed that outcomes for these patients remain poor despite the availability of newer agents, with median overall survival of approximately 8 months. These findings highlight a critical need to develop novel therapies for these patients and also serve as a reference point against which emerging agents for heavily pretreated or highly refractory disease may be evaluated.

Introduction

Treatment of multiple myeloma (MM) has improved during the past decade with the introduction of proteasome inhibitors (PIs) and immunomodulatory drugs (IMiDs) [1–3]. However, long-term outcomes are poor for relapsed patients [4–6]. The natural history of the disease is well chronicled for long-established interventions such as bortezomib (a PI) and lenalidomide and thalidomide (IMiDs). However, there is limited information outside clinical trials on the effects of newer agents on outcomes in relapsed MM patients, such as carfilzomib (PI) and pomalidomide (IMiD) [7].

In 2012, an International Myeloma Working Group (IMWG) study investigated outcomes of patients who were refractory to bortezomib and at least one IMiD (thalidomide, lenalidomide, or both) on the basis of an examination of medical records [7]. The study confirmed that outcomes were poor (median overall survival [OS] of approximately 9 months) in patients who were refractory to agents approved at the time of the analysis. Since the IMWG study was published in 2012, agents that have been approved in the United States and have had an extended period of use in clinical practice include pomalidomide (indicated for treatment of MM patients with two or more prior therapies, including lenalidomide and a PI, and progression within 60 days of last therapy) [8] and carfilzomib (indicated for treatment of MM patients with two or more prior therapies, including bortezomib and an IMiD, and progression within 60 days of last therapy) [9].

In the MM-003 phase III clinical trial examining pomalidomide combined with low-dose dexamethasone in patients with relapsed and refractory MM (at least two previous treatments, including bortezomib and lenalidomide) [10], the pomalidomide treatment group demonstrated significantly longer progression-free survival (PFS) in comparison with those treated with high-dose dexamethasone alone, regardless of previous treatments (4.0 vs. 1.9 months; hazard ratio [HR] = 0.50; p < .001) [11]. Median OS was also significantly improved with the addition of pomalidomide to low-dose dexamethasone in comparison with high-dose dexamethasone (13.1 vs. 8.1 months; HR = 0.72; p = .009) [11].

The phase III FOCUS study compared single-agent carfilzomib to low-dose corticosteroids and optional cyclophosphamide in MM patients who had received three or more prior therapies (including bortezomib, an IMiD [thalidomide or lenalidomide], an alkylating agent, or a corticosteroid), had relapsed on or after at least one prior regimen, and were nonresponsive or had progressed within 60 days of last therapy [12]. Median OS was 10.2 versus 10.0 months for the carfilzomib group and control group, respectively (HR = 0.975; 95% confidence interval [CI], 0.760–1.249; p = .4172), and median PFS was 3.7 versus 3.3 months, respectively.

In light of the availability and use of these new agents for the treatment of relapsed and refractory MM, it is important to understand patient outcomes on the basis of current real-world experience. We analyzed electronic medical records from two independent U.S. databases to determine the median OS of patients with heavily pretreated or double-refractory MM. These analyses also help to assess the need for novel therapies in relapsed disease and to serve as a reference point against which emerging agents for heavily pretreated or highly refractory disease may be evaluated.

Methods

Databases

Medical records of U.S. patients were screened from two independent databases: the IMS LifeLink, IMS Oncology Electronic Medical Records database (IMS Health Inc., Danbury, CT, USA, http://www.imshealth.com/), and the OPTUM database (OPTUM, Inc., Eden Prairie, MN, USA, https://www.optum.com/). Medical records between the index period of 2000 and 2014 were screened within the IMS LifeLink database, whereas the indexing period for the OPTUM database was between 2007 and 2014. These indexing periods were selected on the basis of availability and robustness of data in the respective databases.

Assessments and Analyses

MM treatments considered for this analysis were core drugs (thalidomide, bortezomib, lenalidomide, melphalan, pomalidomide, carfilzomib, and cyclophosphamide), steroids (dexamethasone, prednisone, and prednisolone), and other MM drugs (vincristine, etoposide, doxorubicin, bendamustine, and vorinostat). Regimens were defined as monotherapy or combination therapies (e.g., bortezomib + lenalidomide + dexamethasone). Lines of therapy (LOTs) were identified on the basis of distinct periods associated with each regimen (monotherapy or combination therapy). Subgroup analyses were conducted on those who were only double refractory and on those who were triple/quadruple refractory to PIs and IMiDs. Additionally, median OS from last LOT was assessed for the eligible population and in subgroups from the IMS LifeLink database on the basis of age, Eastern Cooperative Oncology Group (ECOG) performance status at last LOT, or numbers of prior LOTs. Similar subgroup analyses were not feasible on data from the OPTUM database because of the smaller sample size and limited data on clinical variables, particularly ECOG status. OS was defined on the basis of death or loss to follow-up more than 30 days prior to study end date.

Patient Eligibility

Patients with a diagnosis of MM from 2000 to 2011 in the IMS LifeLink database and from 2007 to 2014 in the OPTUM database were eligible for inclusion. International Classification of Diseases–9 codes for MM used in this analysis were 203X, 203.0X, 203.00X, 203.01X, and 203.02X. Patients could not have another cancer diagnosis prior to their diagnosis of MM, other than benign and in situ neoplasms, basal cell carcinoma, and squamous cell carcinoma. Patients were considered eligible if they were refractory to both a PI and an IMiD (criterion 1; double refractory disease) or had received at least three prior LOTs, including a PI and an IMiD, and had shown disease progression within 60 days of completion of their most recent treatment regimen (criterion 2). Patients were defined as refractory if any of three definitions were met (Table 1). Patients who were refractory to three or four prior agents, including both a PI and an IMiD, were considered triple refractory (refractory to two different PIs + one IMiD or one PI + two different IMiDs) or quadruple refractory (refractory to any combination of two different PIs and three different IMiDs), respectively; a patient refractory to the same drug in multiple lines was not considered double or triple/quadruple refractory.

Table 1.

Definitions of refractory status^a

Statistical Analysis

Descriptive measures of the study population were reported. Statistical comparisons between subgroups (age at diagnosis, ECOG score at last LOT, treatment by LOT, and degree of refractoriness) were performed using Kaplan-Meier plot estimates of median OS from the start of last LOT. A p value < .05 was deemed significant.

Results

Patients

A cohort of 4,030 patients with MM was identified from the IMS LifeLink database (Fig. 1A), among whom approximately 90% of patients received an MM diagnosis in 2006 or later. Of the 2,918 patients available for regimen analysis, 500 patients met the eligibility criteria for the study and were further analyzed. Sixty-five percent of eligible patients (n = 323) were double or triple/quadruple refractory to PIs and IMiDs (criterion 1), of whom 253 patients were double refractory, 61 were triple refractory (refractory to a total of any three prior LOTs, including PIs or IMiDs), and 9 were quadruple refractory (refractory to a total of any four prior LOTs, including PIs or IMiDs). The other 35% of patients (n = 177) had received three or more prior LOTs, including a PI and an IMiD (criterion 2) but were not double refractory or triple/quadruple refractory. The median age (range) at MM diagnosis was 66 (31–82) years; at eligibility, the median age (range) was 70 (36–85) years (Table 2). Patients received a median (range) of three (1–25) prior LOTs at eligibility (Table 2), which included carfilzomib (5.8%) and pomalidomide (2.8%; Table 3). A total of 152 (30%) patients had exposure to an alkylating agent (melphalan, cyclophosphamide, or bendamustine) prior to or in their data-eligible LOT.

Figure 1.

Identification of target study population within the IMS LifeLink (A) and OPTUM (B) databases.

Abbreviations: IMiD, immunomodulatory drug; LOT, line of therapy; PI, proteasome inhibitor.

Table 2.

Patient demographics

Table 3.

Treatment regimens received at identification and subsequent lines of therapy

A total of 3,837 patients were screened in the OPTUM database (Fig. 1B), among which approximately 90% of patients were diagnosed with MM after 2009. A total of 162 patients met the eligibility criteria and were analyzed further. Of the 162 patients, 120 patients (74%) were double or triple/quadruple refractory to PIs and IMiDs (met criterion 1); 97, 19, and 4 patients were double, triple, and quadruple refractory, respectively. The other 42 eligible patients (26%) had received three or more prior LOTs, including a PI and an IMiD (only met criterion 2), but were not double or triple/quadruple refractory. The patient demographics of the target study population in the OPTUM database are summarized in Table 2. Median (range) age at diagnosis and eligibility were 66 (35–83) and 67 (39–85) years, respectively. Median (range) of prior LOTs was three (1–8) at eligibility (Table 2), including 5.5% of patients who received any carfilzomib and 3.1% who had received pomalidomide; 2 patients had received both carfilzomib and pomalidomide (Table 3). A total of 66 (41%) patients had exposure to an alkylating agent (melphalan, cyclophosphamide, or bendamustine) prior to or in their data-eligible LOT.

Survival Analyses of Eligible Study Populations and Subgroups

In the IMS LifeLink data set, the median OS was 7.9 months (95% CI, 6.2–9.1; Fig. 2A). The median OS for patients meeting criterion 1 (double refractory) and criterion 2 (three or more LOTs) was 6.7 months and 11.5 months, respectively (p = .1196). In the IMS LifeLink data set, median OS was 7.5 (95% CI, 5.1–8.9) and 5.1 (95% CI, 3.5–8.7) months among double refractory-only (n = 253) and triple/quadruple refractory (n = 70) patients, respectively (Fig. 2B).

Figure 2.

Median overall survival in patients with three or more lines of therapy (including a proteasome inhibitor [PI] and an immunomodulatory drug [IMiD]) or double refractory to a PI and an IMiD (A) and based on refractory status (B) for patients from the IMS LifeLink data set.

Abbreviations: CI, confidence interval; OS, overall survival.

Subgroups of clinical interest within the IMS LifeLink data set were analyzed to determine how median OS varied on the basis of numbers of prior LOT, degree of refractoriness, and patient characteristics. Examination of patients receiving fewer than six prior LOTs versus six or more prior LOTs revealed median OS durations of 9.0 months (95% CI, 6.9–13.0) and 6.2 months (95% CI, 5.0–8.3), respectively (p = .09; Fig. 3A). As was expected, stratification by age revealed that median OS was longer in patients <65 years at eligibility (9.5 months; 95% CI, 7.3–16.3) in comparison with patients ≥65 years (6.7 months; 95% CI, 5.0–8.5; p < .01; Fig. 3B). Similarly, the median OS durations for patients with an ECOG score missing (8.8 months; 95% CI, 7.1–12.6) or ECOG score ≤2 (7.8 months; 95% CI, 4.5–11.9) at last claim were longer than they were for patients with ECOG scores >2 (2.9 months; 95% CI, 1.7–4.9; p < .0001; Fig. 3C). However, there was no significant difference in median OS between females (8.5 months; 95% CI, 5.8–13.0) and males (7.5 months; 95% CI, 5.4–9.0; p = .39).

Figure 3.

Subgroup analyses of median overall survival by number of prior lines of therapy (A), age (B), and Eastern Cooperative Oncology Group status (C) in patients from the IMS LifeLink database.

Abbreviations: CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; LOT, line of therapy; OS, overall survival.

In the OPTUM data set, the median OS was 7.9 months (95% CI, 6.4–10.3; Fig. 4A). Median OS was 7.3 and 10.3 months for patients meeting criterion 1 (double refractory) and criterion 2 (three or more LOTs), respectively (p = .7113). Among double refractory-only patients (n = 97) and triple/quadruple refractory patients (n = 23) in the OPTUM data set, median OS was 8.5 (95% CI, 6.2–11.3) and 3.1 (95% CI, 1.6–13.4) months, respectively (Fig. 4B). Sufficient information on clinical parameters was not available for the OPTUM data set, as it was for the IMS LifeLink data set; thus, comparable subgroup analyses were not performed.

Figure 4.

Median overall survival in patients with three or more lines of therapy (including a proteasome inhibitor [PI] and an immunomodulatory drug [IMiD]) or double refractory to a PI and an IMiD (A) and based on refractory status (B) for patients from the OPTUM data set.

Abbreviations: CI, confidence interval; OS, overall survival.

Median OS was similar between the two data sets (7.9 months in both data sets; p = .5358). Similarly, median OS by refractory status was not significantly different between the IMS LifeLink and OPTUM data sets among double refractory-only patients (p = .8052) or among triple/quadruple refractory patients (p = .6675). On the basis of the similarity in patient demographics and the consistent results observed in the two data sets, the data were combined for additional analyses; the pooled analyses indicated that the median OS was 7.9 months for the total eligible population (n = 662), 7.8 months for patients who were only double refractory (n = 350), and 5.1 months for patients who were triple/quadruple refractory (n = 93) (supplemental online Fig. 1).

Discussion

This study evaluated the treatment landscape of relapsed and refractory MM and characterized outcomes of patients who became refractory to at least one PI and at least one IMiD (double refractory) or were heavily pretreated (three or more LOTs) and progressed within 60 days of the most recent regimen. Analyses of these real-world data from two independent, U.S. patient databases indicated that outcomes remain poor among patients with MM who are heavily pretreated, highly refractory, or both, despite the availability and use of newer PIs and IMiDs, such as carfilzomib and pomalidomide.

As has been discussed previously, in the IMWG study conducted by Kumar et al., MM patients who were refractory to bortezomib and relapsed, refractory, or ineligible for an IMiD were found to show a median OS of 9 months, starting from the date at which patients met the entry criteria [7]. Although this study examined survival from last treatment after having met the entry criteria (criterion 1 or 2), our study also showed that survival outcomes for double refractory-only patients are poor (7.5–8.5 months). It is worth noting that the present study also included approved agents for double refractory MM (carfilzomib and pomalidomide), which were not available at the time of the IMWG study. Additionally, subsequent to study entry, median duration of treatment in this study (more than 3 months) was longer than that reported in the IMWG study (less than 2 months) [7].

Patient characteristics that were associated with shorter median OS in the present study included age 65 years or older and ECOG scores of greater than 2, but these findings must be interpreted with caution because of the descriptive nature of these comparisons. Nevertheless, these observations are in agreement with previous findings that show that good performance status and younger age are associated with longer OS among patients with MM [13, 14]. Together, these findings highlight a continuing unmet need in elderly MM patients and those with poor performance status.

Several limitations of the study should be noted. As with all research conducted with electronic health records or claims data, certain data fields may be subject to inaccurate coding, and the reason for switching of treatments—if it is due to disease progression, refractoriness, or intolerability—is not captured. Furthermore, some fields within the electronic medical records were incomplete for some patients (e.g., missing ECOG status), and some subgroup analyses were conducted with small sample sizes. Despite these limitations, the similarity in OS results in the two separate databases provides a level of confidence in the data.

The efficacies of pomalidomide (as a monotherapy or in combination with dexamethasone) [10, 11, 15] and carfilzomib (monotherapy) [12, 16] in patients who were double refractory to bortezomib and lenalidomide have been previously reported. In the MM-002 phase II trial comparing pomalidomide with or without low-dose dexamethasone, median OS was 13.4 and 12.5 months, respectively, in patients who were refractory to both bortezomib and lenalidomide [15]. A slightly shorter median OS was observed in a subgroup analysis of the larger MM-003 phase III trial, in which patients receiving pomalidomide with low-dose dexamethasone and refractory to both bortezomib and lenalidomide demonstrated a median OS of 11.1 months [10]. Similar median OS was observed with single-agent carfilzomib in heavily pretreated patients who were refractory to both bortezomib and lenalidomide in the PX-171-003-A1 phase II study (11.9 months; 95% CI, 8.4–14.7) [16]. Median OS with carfilzomib in the phase III FOCUS study in patients who had received at least three prior therapies, including bortezomib and an IMiD, and had relapsed on or after at least one prior regimen and were unresponsive or progressed within 60 days of last therapy was 10.2 months [12]. Although these OS improvements for recently approved agents are encouraging in light of the poor survival outcomes associated with double refractory MM (median OS of 9 months) [7], a need for further improvement clearly exists. As such, novel therapies with new mechanisms of action that demonstrate improved efficacy and tolerability in patients with relapsed and refractory MM are urgently needed. Indeed, four new drugs were approved in the United States in 2015, including ixazomib (PI), the histone deacetylase panobinostat inhibitor, and the monoclonal antibodies elotuzumab and daratumumab [17].

In the absence of head-to-head trials, analyses of real-world data may be used as a benchmark for evaluating the potential benefit of novel therapies. The eligibility criteria used to identify patients for the analysis presented here are similar to the inclusion criteria used in a phase II study (SIRIUS) of single-agent daratumumab [18]. Daratumumab is a first-in-class, human IgG1 monoclonal antibody that targets CD38, a glycoprotein that is highly expressed in MM cells [19–21]. In the SIRIUS phase II trial—which examined the activity of daratumumab monotherapy (16 mg/kg) in patients with MM who had received three or more prior LOTs, including a PI and an IMiD, or who were double refractory to a PI and IMiD—median OS was 17.5 months (95% CI, 13.7 to not estimable) [18]. A naïve comparison of the OS Kaplan-Meier curve from the SIRIUS study (17.5 months) and the pooled analysis of both data sets (7.9 months) presented here suggests a survival benefit with daratumumab versus the real-world historical control (difference of 10.2 months) [22]. Limitations of this type of naïve comparison include differences in patient populations, comparison of randomized clinical trial data with observational data, and differences in study period (resulting in variability in available standard of care).

Conclusion

Although newer PIs and IMiDs, such as carfilzomib and pomalidomide, have been introduced into the treatment regimen, our study of real-world data from electronic medical records of two independent U.S. databases suggests that median OS durations remain poor (approximately 8 months) in patients with MM who are heavily pretreated, those refractory to a PI and an IMiD, or both. These data highlight the critical need for new treatments for this patient population and also provide a reference point against which novel agents for advanced MM may be evaluated. A comparison of the median OS observed in the IMS LifeLink and OPTUM data sets with a study of daratumumab monotherapy suggests a survival benefit with daratumumab in patients with heavily pretreated MM, highly refractory MM, or both.

See http://www.TheOncologist.com for supplemental material available online.

Author Contributions

Conception/design: Saad Usmani, Tahamtan Ahmadi, Yvette Ng, Annette Lam, Avinash Desai, Ravi Potluri, Maneesha Mehra

Provision of study material or patients: Ravi Potluri, Maneesha Mehra

Collection and/or assembly of data: Yvette Ng, Annette Lam, Ravi Potluri, Maneesha Mehra

Data analysis and interpretation: Saad Usmani, Tahamtan Ahmadi, Yvette Ng, Annette Lam, Avinash Desai, Ravi Potluri, Maneesha Mehra

Manuscript writing: Saad Usmani, Tahamtan Ahmadi, Yvette Ng, Annette Lam, Avinash Desai, Ravi Potluri, Maneesha Mehra

Final approval of manuscript: Saad Usmani, Tahamtan Ahmadi, Yvette Ng, Annette Lam, Avinash Desai, Ravi Potluri, Maneesha Mehra

Disclosures

Saad Usmani: Amgen/Onyx, Celgene, Sanofi, Takeda (C/A), Array Biopharma, Takeda, Sanofi, Janssen, Pharmacyclics, Onyx (RF), Onyx, Celgene, Takeda (H); Tahamtan Ahmadi: Janssen Research and Development (E, OI); Yvette Ng: Johnson & Johnson (E, OI); Annette Lam: Janssen (E, OI); Avinash Desai: Janssen (E); Ravi Potluri: SmartAnalyst, Inc., contracted by Janssen Global Services (E); Maneesha Mehra: Janssen (E).

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board