Real-World Data on Clinical Features, Outcomes, and Prognostic Factors in Multiple Myeloma from Miyazaki Prefecture, Japan

The prognosis of multiple myeloma (MM) has improved with the introduction of novel agents. These data are largely derived from clinical trials and might not reflect real-world patient outcomes accurately. We surveyed real-world data from 284 patients newly diagnosed with MM between 2010 and 2018 in Miyazaki Prefecture. The median follow-up period was 32.8 months. The median age at diagnosis was 71 years, with 68% of patients aged >65 years. The International Staging System (ISS) stage at diagnosis was I in 18.4% of patients, II in 34.1%, and III in 47.5%. Bortezomib-containing regimens were preferred as initial treatment; they were used in 147 patients (51.8%). In total, 80% of patients were treated with one or more novel agents (thalidomide, lenalidomide, or bortezomib). Among 228 patients who were treated with novel agents as an initial treatment, the overall response rate (partial response (PR) or better) to initial treatment was 78.4%, and the median time to next treatment (TTNT) was 11.6 months. In the multivariate analysis, PR or better responses to initial treatment were independently favorable prognostic factors for TTNT. The median survival time after initial therapy for patients with novel agents was 56.4 months and 3-year overall survival (OS) was 70.4%. In multivariate analysis, ISS stage I/II disease and PR or better response to initial treatment, and autologous stem cell transplantation (ASCT) were identified as independent prognostic factors for overall survival (OS).


Introduction
Multiple myeloma (MM) is a disease characterized by the proliferation of malignant plasma cells. Clinical features include the so-called CRAB criteria (hyperCalcemia, Renal 2 of 9 dysfunction, Anemia, Bone lesions) [1]. The prognosis of patients with MM has improved during the past two decades [2]. The introduction of "novel agents" such as thalidomide, lenalidomide, and bortezomib as well as autologous stem cell transplantation (ASCT) has contributed to this improvement, but MM remains incurable with current therapy.
Data demonstrating improvement in MM outcomes are largely derived from clinical trials and might not reflect "real-world" patient outcomes accurately. Several studies based on real-world data have been reported, which are useful for selecting and optimizing treatment in clinical practice [2][3][4][5][6][7][8][9][10]. Studies based on real-world data may be divided into two groups: studies at tertiary referral centers and population-based studies. We conducted a chart review in order to survey the clinical features, therapy options, and outcomes of newly diagnosed MM between 2010 and 2018 in Miyazaki Prefecture, Japan.

Patients and Data Collection
We retrospectively surveyed clinical data of patients newly diagnosed with MM between January 2010 and April 2018 at 7 hospitals in Miyazaki Prefecture, Japan. The ethical approval for this study was obtained from the Research Ethics Committee of the Faculty of Medicine, University of Miyazaki.

Definitions and Clinical Outcome Variables
Clinical staging was based on the International Staging System (ISS) [11]. Standardrisk chromosomal abnormalities (CA) by interphase fluorescence in situ hybridization was defined as absence of del (17p), t (4; 14), and t (14;16), whereas high-risk CA was defined as presence of at least one of these abnormalities [12]. Response was based on the International Uniform Response Criteria [13], which includes the following categories: stringent complete response (sCR), complete response (CR), very good partial response (VGPR), partial response (PR), stable response (SD), and progressive disease (PD). Overall response was defined as sCR + CR + VGPR + PR. Overall survival (OS) was calculated from the day of initial therapy to the date of death or last observation. Patients who remained alive at the time of the last follow-up were censored.

Statistical Analysis
Factors that might influence response or survival were analyzed using logistic regression. Categorical variables were first analyzed in univariate analyses. Variables that were significantly associated with response or survival in univariate analyses were included in a Cox proportional hazards model for multivariate analysis. Statistical significance was defined as p < 0.05. The Kaplan-Meier method was used to estimate OS. The log-rank test was used to compare OS between groups. All data were analyzed using EZR (Easy R) for the R commander, version 1.37.

Patients Characteristics at Diagnosis and Initial Treatments
Between January 2010 and April 2018, there were 284 patients diagnosed with MM at 7 hospitals in Miyazaki Prefecture, Japan. Table 1   17 26 a , Standard risk defined as absence of high risk chromosomal abnormalities; b , High risk chromosomal abnormalities defined as presence of del 17p and/or t (4;14) and/or t (14;16); ASCT, autologous stem cell transplantation; CR, complete response; ISS, International Staging System; MP, melphalan and prednisone; NA, not available; OR, overall response; PD, progressive disease; PR, partial response; sCR, stringent complete response; SD, stable disease; VGPR, very good partial response. p # , p value between patient treated with novel agent and patients treated without novel agent; p ☥ , p value between patient aged ≤65 years and patients aged >65 years.
As initial treatment, 37 patients were treated with bortezomib + lenalidomide + dexamethasone, 147 with other bortezomib-containing regimens, 26 with lenalidomide-containing regimens, 18 with thalidomide-containing regimens, and 56 with conventional chemotherapy. The details of conventional chemotherapy are as follows: melphalan + prednisone was used in 37 patients, vincristine + adriamycin + dexamethasone in 6 patients, high-dose dexamethasone in 5 patients, prednisone in 1 patient, and palliative radiation therapy in 7 patients. Overall, 80% of patients were treated with one or more novel agents as bortezomib, lenalidomide, or thalidomide. ASCT was performed in 47 patients , p value between patient aged ≤65 years and patients aged >65 years.
As initial treatment, 37 patients were treated with bortezomib + lenalidomide + dexamethasone, 147 with other bortezomib-containing regimens, 26 with lenalidomide-containing regimens, 18 with thalidomide-containing regimens, and 56 with conventional chemotherapy. The details of conventional chemotherapy are as follows: melphalan + prednisone was used in 37 patients, vincristine + adriamycin + dexamethasone in 6 patients, high-dose dexamethasone in 5 patients, prednisone in 1 patient, and palliative radiation therapy in 7 patients. Overall, 80% of patients were treated with one or more novel agents as bortezomib, lenalidomide, or thalidomide. ASCT was performed in 47 patients (16.5%).
The following analysis was focused on 228 patients who were treated with novel agents as an initial treatment. The median follow-up period was 32.8 months.

The Best Response to Initial Treatment
The best response to initial treatment was sCR in 11 patients, CR in 17 patients, VGPR in 49 patients, PR in 68 patients, SD in 29 patients, and PD in 11 patients. Among patients who underwent ASCT, 18.9% achieved CR or better response before ASCT.
The overall response rate, defined as the proportion of patients with PR or better response, was 78.4%. In univariate analysis, age ≤ 65 years (p = 0.019) was associated with favorable response.

Time to Next Treatment (TTNT) and 2nd Line Regimens
The median TTNT was 11.6 months for patients treated with novel agents (Figure 1a). TTNT in patients aged ≤65 years was significantly superior to that in patients aged >65 (p = 0.011) (Figure 1b). The median TTNT in patients with stage I, II, and III was 10.5, 14.2, and 9.3 months, respectively. TTNT was longer in patients with stage I/II disease compared with those with stage III disease (Figure 1c).
(16.5%). The following analysis was focused on 228 patients who were treated with novel agents as an initial treatment. The median follow-up period was 32.8 months.

The Best Response to Initial Treatment
The best response to initial treatment was sCR in 11 patients, CR in 17 patients, VGPR in 49 patients, PR in 68 patients, SD in 29 patients, and PD in 11 patients. Among patients who underwent ASCT, 18.9% achieved CR or better response before ASCT.
The overall response rate, defined as the proportion of patients with PR or better response, was 78.4%. In univariate analysis, age ≤ 65 years (p = 0.019) was associated with favorable response.

Time to Next Treatment (TTNT) and 2nd Line Regimens
The median TTNT was 11.6 months for patients treated with novel agents ( Figure  1a). TTNT in patients aged ≤65 years was significantly superior to that in patients aged >65 (p = 0.011) (Figure 1b). The median TTNT in patients with stage I, II, and III was 10.5, 14.2, and 9.3 months, respectively. TTNT was longer in patients with stage I/II disease compared with those with stage III disease (Figure 1c). The second line regimens in patients treated with novel agents as an initial treatment were summarized in Table 2. Among patients treated with triplet regimens, 15 (71.4%) patients were secondary treated with next-generation novel agents, such as daratuzumab, elotuzumab, carfilzomib, ixazomib, or pomalidomide. Among patients with bortezomib, 35 (39.3%) and 19 (21.3%) were secondary treated with lenalidomide and next-generation novel agents, respectively. Among patients with lenalidomide, 5 (50.0%) were secondary treated with bortezomib. The second line regimens in patients treated with novel agents as an initial treatment were summarized in Table 2. Among patients treated with triplet regimens, 15 (71.4%) patients were secondary treated with next-generation novel agents, such as daratuzumab, elotuzumab, carfilzomib, ixazomib, or pomalidomide. Among patients with bortezomib, 35 (39.3%) and 19 (21.3%) were secondary treated with lenalidomide and next-generation novel agents, respectively. Among patients with lenalidomide, 5 (50.0%) were secondary treated with bortezomib.

Factors Which Affected on TTNT and OS
Among patients who were treated with novel agents as an initial treatment, both TTNT and OS were significantly better in the deeper response groups by response to initial treatment. TTNT in patients achieving PR or better response was 15.3 months, whereas TTNT of patients achieving SD or PD was 6.6 months (p < 0.001) (Figure 3a). MST in patients achieving PR or better response was 68.2 months, whereas MST of patients achieving SD or PD was 39.6 months. Patients achieving PR or better response after the initial treatment had better survival than patients with SD or PD (p < 0.001) (Figure 3b).

Factors which Affected on TTNT and OS
Among patients who were treated with novel agents as an initial treatment, both TTNT and OS were significantly better in the deeper response groups by response to initial treatment. TTNT in patients achieving PR or better response was 15.3 months, whereas TTNT of patients achieving SD or PD was 6.6 months (p < 0.001) (Figure 3a). MST in patients achieving PR or better response was 68.2 months, whereas MST of patients achieving SD or PD was 39.6 months. Patients achieving PR or better response after the initial treatment had better survival than patients with SD or PD (p < 0.001) (Figure 3b).

Factors which Affected on TTNT and OS
Among patients who were treated with novel agents as an initial treatment, both TTNT and OS were significantly better in the deeper response groups by response to initial treatment. TTNT in patients achieving PR or better response was 15.3 months, whereas TTNT of patients achieving SD or PD was 6.6 months (p < 0.001) (Figure 3a). MST in patients achieving PR or better response was 68.2 months, whereas MST of patients achieving SD or PD was 39.6 months. Patients achieving PR or better response after the initial treatment had better survival than patients with SD or PD (p < 0.001) (Figure 3b). (a) TTNT in those who achieved PR/better versus those who did not. (b) OS in those who achieved PR/better versus those who did not. CR, complete response; OS, overall survival; PD, progressive disease; PR, partial response; sCR, stringent complete response; SD, stable disease; TTNT, time to next treatment; VGPR, very good partial response. Figure 3. Time to next treatment (TTNT) and overall survival (OS) by response to initial therapy. (a) TTNT in those who achieved PR/better versus those who did not. (b) OS in those who achieved PR/better versus those who did not. CR, complete response; OS, overall survival; PD, progressive disease; PR, partial response; sCR, stringent complete response; SD, stable disease; TTNT, time to next treatment; VGPR, very good partial response.
We analyzed the effect of initial therapy response and ASCT together with patient characteristics. In univariate analyses, age ≤ 65 year, ISS stage I/II stage, PR or better response to initial treatment, and ASCT were associated with longer TTNT, and age ≤65 years, ISS stage I/II disease, PR or better response to initial treatment, and ASCT were associated with favorable OS (Tables 3 and 4). In multivariate analysis, PR or better responses to initial treatment were independently favorable prognostic factors for TTNT, and ISS stage I/II disease and PR or better response to initial treatment, and ASCT were independent favorable prognostic factors for OS.

Discussion
We retrospectively surveyed patient characteristics, clinical features, treatment options, and outcomes in patients newly diagnosed with MM. These data provide valuable information on the circumstances of MM and real-world outcomes following treatment and factors that influence prognosis.
The median age at diagnosis of our patients was 71 years, and 68% of patients in this study were aged >65 years. These results were similar to those in previous realworld reports from a European multicenter observational study and a Japanese regional study [3,4], but the median age of our patients was approximately 5 years older than in previous reports from the Mayo Clinic, a Japanese Society of Myeloma (JSM) multicenter study, and Czech registry data [2,5,6]. In our cohort, only 18.4% of patients had ISS stage I disease at diagnosis. This proportion was similar to the proportion in the European observational study (16%), but was much lower than that in reports from the Mayo Clinic (30%), JSM multicenter study (26.5%), and Czech registry data (31.9%) [2,[4][5][6]. Older age was correlated with higher ISS stage, with 10.9% of patients aged >65 years presenting with ISS stage I disease in our cohort. This was also observed in the Czech registry data, where only 12% of patients aged ≥80 years presented with ISS stage I disease [5]. The lower proportion of patients with ISS stage I disease in both our cohort and the European observational study might be due to the higher proportion of elderly patients compared with other reports.
Current guidelines recommend initial treatment with one or more novel agents and ASCT for eligible patients. In our cohort, about 80% of patients received one or more novel agents as the initial treatment. In contrast, only 16.5% of patients underwent ASCT, which was much lower than that in the Mayo Clinic reports or JSM multicenter study, in which 37% and 34% of patients received ASCT, respectively [2,6]. One reason for this difference might be the larger proportion of elderly patients in our cohort. In fact, this gap narrowed when the analysis was restricted to patients aged ≤65 years; 40.0% and 56% of patients aged ≤65 years received ASCT in our cohort and the Mayo Clinic cohort, respectively, although the gap was still present. Another major reason was the difficulty in accessing transplant facilities.
Among patients who were treated with novel agents as an initial therapy, most (78.4%) achieved PR or better response after initial treatment. The proportion of patients who achieved CR (9.2%) and VGPR or better response (41.6%) in our cohort was much lower than that in the European observational study, in which 32% and 74% of patients achieved CR and VGPR or better response, respectively [4]. In addition, the TTNT in our cohort was 11.6 months, which was also inferior to the corresponding value (progression-free survival: 16.3 months) in the European observational study. We are not able to account for this difference of a deeper response (CR/VGPR) and TTNT between our cohort and the European observational study at this time. However, more elderly patients in our cohort might account for the difference. Even though the response rate in our cohort was lower and TTNT was shorter than those reported in other real-world studies, OS seemed to be similar. MST in our cohort was 56.4 months, which was comparable with MST of 50.3 months in the Czech registry data and 51.8 months in a previous Japanese regional study [3,5], but was inferior to the 6.1 years or 60.6 months in reports from the Mayo Clinic or JSM multicenter study [2,6]. In univariate analysis, age ≤ 65 years, ISS stage I/II disease, PR or better response to initial treatment, and ASCT were associated with favorable prognosis. The inferior OS in our cohort compared to reports from the Mayo Clinic or JSM multicenter study might be due to the higher proportion of elderly patients, lower proportion of patients with ISS stage I disease, and lower proportion of patients who underwent ASCT. Patient characteristics in our cohort and outcomes resembled the results of the European observational study and Czech registry data, which might reflect real-world practice for MM.
In addition to patient age, disease burden, and treatment choice, which are all based on pre-treatment status, PR or better response to initial treatment was associated with longer TTNT and better OS, which was consistent with many previous reports [6,14]. This initial treatment effect was also observed with multivariate analysis.
This study had some limitations. It was retrospective in design. Each physician determined the choice of treatment, and treatment doses and durations varied across individuals. Physician definitions of treatment response may not be as rigorous as the evaluation criteria used in clinical trials. However, our data constituted a population-based cohort. Therefore, the results of this study reflect daily practice in the era of novel agents. Finally, early identification of patients with poor prognosis after the initial treatment will allow us to change the therapeutic strategy earlier, which may contribute to better survival.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.