Introduction

Multiple myeloma (MM) is the second most common hematological malignancy and accounts for roughly 10% of all hematological malignancies [1]. The United States had an estimated lifetime incidence of MM at 0.76% in 2022, with 34,470 new cases reported annually [2]. Current treatment guidelines recommend upfront autologous stem cell transplantation (ASCT) for MM transplant-eligible patients, especially for those with high-risk cytogenetics [1, 3]. However, studies have shown that frontline ASCT primarily improves progression-free survival (PFS) rather than overall survival (OS) [4]. In the IFM 2009 trial [5], patients who received bortezomib, lenalidomide, and dexamethasone (VRD) followed by upfront ASCT demonstrated superior PFS (p < 0.001) compared to the non-ASCT group, but OS was comparable (p = 0.87). The recent DETERMINATION trial yielded similar results by comparing VRD alone with lenalidomide maintenance and VRD followed by ASCT with lenalidomide maintenance [6]. To date, several studies have addressed outcomes for MM patients who have undergone ASCT. However, the risk factors for poor outcomes vary between studies. Tandon et al. analyzed the outcomes of MM patients who received frontline treatment and found that patients with high-risk cytogenetics had poorer PFS and OS [7]. Factors such as response ≥ very good partial response (VGPR), creatinine ≥ 2 mg/dl and age were not significant. Recently, a large retrospective study reported that pre–ASCT treatment response and age can predict outcomes [8]. Another study evaluated the risk factors that impact post–ASCT outcomes for MM patients based on the Center for International Blood and Marrow Transplant Research (CIBMTR) database, finding that patients with treatment response ≥ VGPR, Karnofsky performance score ≥ 90, International Staging System (ISS) III compared to I–II, and high-risk cytogenetics were significant factors for post–ASCT PFS and OS prediction [9]. Moreover, post–ASCT measurable residual disease (MRD) has been shown to predict outcomes and guide post–transplant management in several studies [10,11,12]. However, routine MRD evaluation is limited due to financial constraints and availability. Thus, there is a need for affordable and easily accessible predictive factors for post–ASCT outcomes.

Materials and methods

Study population

This is a retrospective study. Patients at Taipei Veterans General Hospital newly diagnosed with MM according to International Myeloma Working Group (IMWG) criteria were enrolled in this study [13]. We screened MM patients diagnosed between October 1, 2006 and August 31, 2020. Patients who did not receive induction treatment followed by frontline ASCT were excluded. Institutional review board approval was obtained at Taipei Veterans General Hospital (no. 2022-05-001AC).

Induction treatment, stem cell harvesting, and autologous stem cell transplantation

Induction treatment varies based on the timing of diagnosis and the availability of different treatment options. Chemotherapy-based regimens commonly include vincristine, doxorubicin, and dexamethasone (VAD), while novel agent-based regimens often use bortezomib, thalidomide, and dexamethasone (VTD).

All patients received stem cell mobilization with chemotherapy and granulocyte colony–stimulating factor (G-CSF) [14, 15]. Peripheral blood stem cell harvesting started when white cell count (WBC) was ≥ 1,000/µl and peripheral blood hematopoietic progenitor cells were ≥ 20 × 106 cells/L [16, 17]. Leukapheresis was performed using the COBE Spectra apheresis system (Version 6.1, COBE Laboratories) with the processed volume of 10,000 ml or 2–3 times of total blood volume, depending on the patients’ body weight and age [18]. The collected CD34+ hematopoietic stem cell count was measured using flow cytometry. Before use, stem cells were preserved in liquid nitrogen freezers (MVE HEco 1500Series Freezer) [18]. The detailed apheresis procedure is outlined in our previous study [19].

Conditioning chemotherapy with melphalan was administered before ASCT [20]. Melphalan dosage depended on the patients’ age, Eastern Cooperative Oncology Group (ECOG) performance status, and the treating physician’s discretion [20]. On the day of ASCT, the cryopreserved stem cells were thawed in a thermostatic water bath (37 °C) and reinfused via a central venous catheter. A small number of thawed stem cells was sent for cell viability analysis and culture for quality control. G-CSF was prescribed after ASCT and until neutrophil engraftment. The treating physician decided on the starting day of G-CSF [21, 22].

Data collection

Patient data were collected at multiple timepoints, including at MM diagnosis, before ASCT, on the day of ASCT, and after ASCT. Baseline information encompassed sex, ISS disease stage, ECOG performance status, disease characteristics, bone marrow pathology, comorbidities, induction treatment regimens, and laboratory data obtained at diagnosis. Extramedullary disease was recorded and defined as hematologenous spreading of MM according to an expert review conducted in 2021 [23]. Paraosseous plasmacytoma was not included in this definition. The total number of CD34+ cells collected on the first stem cell harvesting was also recorded. Before ASCT, defined as within one week before stem cell infusion, we collected conditioning melphalan dosage, light chain ratio, and treatment response according to IMWG response criteria [24]. On the day of ASCT, patients’ age at ASCT and the duration between stem cell mobilization and ASCT were collected. After ASCT, we collected laboratory data on day + 15 (D15), including absolute monocyte counts (AMC), absolute neutrophil counts (ANC), absolute lymphocyte counts (ALC), and platelet counts (PLT). Additionally, information on post–ASCT bortezomib maintenance, cytomegalovirus (CMV) infection within one year after ASCT, and fungal infection within one year after ASCT was recorded.

Statistical analysis

Patients’ baseline information was categorized and presented as the total number (n) and proportion (%) for further analysis. The primary endpoint of this study was PFS. The secondary endpoint was OS. Patients lost to follow-up were censored. For patients who received ASCT followed by allogeneic hematopoietic stem cell transplantation (alloHSCT), the last follow-up date was considered as day 0 of alloHSCT. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards models to analyze the impact of each factor on survival. A univariate Cox proportional hazards model identified potential risk factors for decreased survival, with those having p < 0.1 proceeding to backward multivariate Cox regression analysis. The cumulative incidence of disease progression and survival were calculated using the Kaplan–Meier method. The relationship between age ≥ 65 and melphalan dosage ≥ 180 mg/m2 was evaluated using a Chi-square test. SAS 9.4 (SAS Institute Inc., Cary, NC) and STATA 15.1 (StataCorp, College Station, TX) were used for data management and statistical analysis. Statistical significance was defined as p < 0.05.

Results

Clinical characteristics of the study population

A total of 628 MM patients at Taipei Veterans General Hospital diagnosed between October 2006 and August 2020 were identified. After exclusions, the final study population consisted of 150 MM patients who underwent upfront ASCT (shown in Fig. 1). The median ASCT age of the patients was 58 (range, 24–75), with 34 patients (22.7%) aged ≥ 65 and 116 patients (77.3%) younger than 65 years. Among the study population, 54.7% were men. ECOG performance status at MM diagnosis with a score of 0–1 was reported in 115 patients (76.7%) and ≥ 2 in the remaining patients (23.3%). ISS stages I & II, and III were 74.7%, and 24.0%, respectively. Extramedullary disease was present in 25 cases (16.7%). Comorbidities including heart failure, chronic pulmonary disease, diabetes mellitus, and hypertension were reported in 2.0%, 4.0%, 7.3%, and 8.7% of the cases, respectively. Induction therapy consisted of VAD in 6.7% of the patients, VTD in 46.0%, bortezomib, cyclophosphamide, thalidomide, and dexamethasone (VCTD) in 14.0%, VCD in 10.7%, and other regimens in 22.7% of the patients. Treatment response ≥ VGPR before ASCT was identified in 125 patients (83.3%). A conditioning melphalan dosage of less than 140 mg/m2 was reported in 31 patients (20.7%), between 140 and 179 mg/m2 in 40 patients (26.7%), and ≥ 180 mg/m2 in the majority of the study population (79 cases, 57.2%). In the context of ASCT, conventional melphalan doses are 200 mg/m2, 140 mg/m2, and 100 mg/m2. Most of the conditioning dosages in our study deviated slightly from these standard targets due to various clinical situations (dosage adjustments for non-integer dosages, cost-effectiveness considerations, etc.). Furthermore, certain clinicians opted to reduce the intended conditioning dosage by 10–20%, considering the patient’s clinical condition, resulting in a dosage between 140 and 179 mg/m2. In patients aged ≥ 65, conditioning melphalan dosage ≥ 180 mg/m2 was 29.4%, and the percentage was significantly higher in the group aged < 65 (59.5%) (p = 0.002). Post–ASCT D15 laboratory data showed ANC > 2,000/µl in 66.7% of patients, ALC > 500/µl in 69.3%, AMC > 500/µl in 88.0%, and PLT > 80,000/µl in 49.3%. Bortezomib maintenance was given to 45.3% of patients, while CMV infection was reported in 4.7% and fungal infection in 2.0%. Detailed information on the patients’ characteristics is listed in Table 1.

Fig. 1
figure 1

Patient selection flowchart

Full size image
Table 1 Baseline characteristics of myeloma patients receiving front-line autologous stem cell transplantation
Full size table

Risk factors for disease progression after autologous stem cell transplantation

The median PFS was 2.9 years, with the cumulative incidence shown in Fig. 2. Among the variables, those with a p value of < 0.1 in the univariate analysis were identified as possible risk factors associated with PFS and underwent multivariate analysis (Table 2). In the backward multivariate analysis, ASCT age ≥ 65 (HR 1.94, 95% CI 1.08–3.47) and the presence of extramedullary disease (HR 2.53, 95% CI 1.53–4.19) were factors negatively associated with PFS. On the other hand, treatment response ≥ VGPR before ASCT (HR 0.52, 95% CI 0.31–0.87) and total CD34+ cells collected ≥ 4 × 106 cells/kg on the first stem cell harvesting (HR 0.52, 95% CI 0.32–0.87) were significant predictors positively associated with PFS after ASCT (Table 2).

Fig. 2
figure 2

Kaplan–Meier survival curves of progression-free survival and overall survival

Full size image
Table 2 Risk factors for disease progression of myeloma patients receiving front-line autologous stem cell transplantation
Full size table

Risk factors for decreased survival after autologous stem cell transplantation

Variables with a p < 0.1 in the univariate analysis were selected for the multivariate analysis (Table 3). In the backward multivariate Cox regression model, ISS stage III, compared to stage I & II (HR 2.06, 95% CI 1.05–4.04), the presence of extramedullary disease (HR 3.92, 95% CI 2.03–7.58), light chain ratio ≥ 100 before ASCT (HR 7.08, 95% CI 1.45–34.59), post–ASCT CMV infection (HR 9.43, 95% CI 3.09–28.84), and a lower conditioning melphalan dose (< 140 mg/m2; HR 2.75, 95% CI 1.23–6.17) were factors associated with a shorter OS. In contrast, D15 AMC > 500/µl (HR 0.36, 95% CI 0.17–0.79) and D15 PLT > 80,000/µl (HR 0.48, 95% CI 0.24–0.94) were significant factors positively associated with overall survival (Table 3). The Kaplan–Meier estimator depicted the cumulative probability of the death curve (Fig. 2), with a median overall survival of 7.0 years.

Table 3 Risk factors for decreased survival of myeloma patients receiving front-line autologous stem cell transplantation
Full size table

Discussion

In this study, we observed that age ≥ 65 at ASCT and the presence of extramedullary disease were associated with shorter PFS. Conversely, treatment response ≥ VGPR before ASCT and total CD34+ cells collected ≥ 4 × 106 cells/kg on the first stem cell harvesting correlated with prolonged time to disease progression. Factors impacting survival were further analyzed, revealing that extramedullary disease, light chain ratio ≥ 100 before ASCT, post–ASCT CMV infection, a lower conditioning melphalan dose (< 140 mg/m2 compared with ≥ 180 mg/m2), and ISS stage III at MM diagnosis were linked to shorter survival. In contrast, D15 AMC > 500/µl and D15 PLT > 80,000/µl were associated with extended survival. Notably, D15 AMC and PLT were identified as predictors of improved OS for MM patients undergoing ASCT, which is a novel finding. Additionally, extramedullary disease corresponded to shorter PFS and OS.

To the best of our knowledge, this is the first study describing D15 PLT as a predictive factor for survival. Some studies have reported that post–ASCT PLT recovery can affect outcomes, but not as early as our finding. We also identified a trend toward better PFS in patients with D15 PLT > 80,000/µl. Ninan et al. found that post–ASCT PLT ≥ 150,000/µl was associated with better OS (p < 0.001), with a shorter time to reach PLT > 50,000/µl in this group (17.2 days vs. 26.5 days, p < 0.001) [25]. This result reveals that early PLT recovery predicts better OS in patients who have undergone ASCT. A recent nationwide Japanese study evaluated the impact of the delay of platelet recovery after ASCT, defined as day + 28 PLT < 50,000/µl or PLT transfusion dependence, on survival in non-Hodgkin lymphoma and MM patients. They found that both lymphoma and MM patients with delayed PLT recovery had worse OS and PFS. Furthermore, they reported that, in addition to total CD34+ cells infused, patient factors including disease status at ASCT and ISS stage can affect PLT recovery, implying that the bone marrow microenvironment may have a role in PLT recovery. Previous studies have shown that impaired bone marrow microenvironment can impact PLT recovery and bone marrow function [26,27,28]. It has also been reported that megakaryocytes have a supportive role in expanding hematopoietic stem cells, confirmed in murine models [29, 30]. Collectively, the predictive value of D15 PLT > 80,000/µl may reflect a better bone marrow function, which is correlated with an improved OS. Similarly, several studies have shown that CD34+ yielded at a dose of 4 × 106 cells/kg or higher after stem cell mobilization improves post–ASCT survival in MM patients, probably due to the relatively healthy bone marrow stem cell niches and microenvironment [31,32,33,34].

Early monocyte recovery is another novel finding in our work. Although there is some evidence addressing the early recovery of monocytes conferring better survival, no studies have reported results as early as ours. Turcotte et al. reported that early recovery of monocytes on day + 28 in patients undergoing hematopoietic stem cell transplantation is linked to longer survival [35]. Similarly, Dhakal et al. found that MM patients who received alloHSCT and had AMC > 300/µl at day + 60 and + 100 exhibited improved OS [36]. DeCook et al. observed that AMC > 300/µl at day + 100 after alloHSCT led to improved OS in myeloid and lymphoid malignancies [37]. These studies reveal that early AMC recovery is correlated with prolonged survival, and the benefits of early AMC recovery can be applied to different types of hematological malignancies after transplantation. Binder et al. used AMC and ALC as biomarkers of immune recovery in MM patients [38]. Immune reconstitution at post–MM treatment one month, defined as the recovery of AMC and ALC to normal reference range, correlated with significantly longer survival (HR 0.63, 95% CI 0.50–0.80, p < 0.001). Most patients with immune dysregulation at diagnosis had monocytopenia and increased to the normal range (> 500/µl) after immune recovery. In brief, early AMC recovery may reflect early immune reconstruction and better bone marrow function, thus prolonging survival [39].

Our data suggests poorer PFS and significantly worse OS in the lower-dose melphalan group, corroborating previous studies showing improved outcomes with higher melphalan doses [9, 40]. Despite age-related effects observed in the univariate analysis, they weren’t considered in the multivariate analysis due to model selection limitations imposed by the small sample size.

The presence of extramedullary disease significantly diminished both PFS and OS in this study, aligning with findings from prior investigations [41, 42]. We can also confirm that advanced ISS stage was predictive of poor outcomes after ASCT [43,44,45,46,47,48]. Additionally, MM patients with high light chain ratio may have more aggressive disease and a lower VGPR rate [49]. Several studies have also reported the correlation between treatment response before ASCT and post–ASCT outcomes [50,51,52]. These results are similar to those in our study. Another study conducted by Brioli et al. aimed to evaluate the relationship between treatment response before ASCT and patient outcomes [53]. They found that a response of ≥ VGPR before ASCT was the strongest predictor for better OS and PFS. However, following ASCT, VGPR failed to predict PFS in high-risk and ultra-high-risk patients identified through fluorescence in situ hybridization (FISH). Even after induction, VGPR did not correlate with improved PFS in ultra-high-risk patients. These observations lead to the conclusion that treatment response and cytogenetic risk features exert independent influences on outcomes. Additionally, it was noted that a favorable treatment response does not necessarily translate into better outcomes for high-risk and ultra-high-risk patients.

Lenalidomide is the only category 1 recommended post–ASCT maintenance therapy for MM, supported by clinical trials [54, 55]. In our study, bortezomib demonstrated significant PFS and OS improvement in the univariate analysis, aligning with documented efficacy in studies, although it was not selected for multivariate analysis, probably due to the small sample size [56,57,58]. These findings suggest that bortezomib can be a viable option for maintenance therapy in MM patients after ASCT, expanding the available choices beyond lenalidomide.

Our findings reveal a higher risk of early death associated with post–ASCT CMV infection. Despite guidelines recommending selective CMV monitoring, reactivation is not uncommon after ASCT [59,60,61,62,63]. In our institution, CMV monitoring focuses on critical cases. Our results imply that CMV infection should be kept in mind after ASCT, although regular monitoring may be unnecessary.

This retrospective study has limitations. Variations in induction therapy, conditioning melphalan dosage, and post–ASCT maintenance therapy were observed. The considerable variability in treatment regimens stems from the extended duration of treatment in our study group and the ongoing evolution of novel agents. Various induction regimens may potentially influence our results. Additionally, the FISH-based high-risk cytogenetic testing was limited because the test was not covered by Taiwan National Health Insurance. However, these limitations reflect real-world MM management and diverse patient populations. The main maintenance regimen was bortezomib due to coverage restrictions on lenalidomide by Taiwan National Health Insurance, which may affect risk factors. Only 10% of our patients received lenalidomide maintenance. Nonetheless, our data suggests that bortezomib can be a candidate of viable maintenance therapy options. The prevalence of D15 AMC > 500/µl in 88% of patients restricts its utility as a prognostic marker for reduced survival in MM, as it would only identify a minority (12%) percentage of patients with an unfavorable prognosis. Lastly, our sample size was small, possibly reducing statistical power and masking certain prognostic factors. Some of our factors were significant in the univariate analysis but were not selected for multivariate analysis by model selection. Larger studies are needed to address this issue.

Conclusion

We discovered previously unreported predictors of better OS after ASCT in MM patients: post–ASCT D15 AMC > 500/µl and PLT > 80,000/µl. Additionally, our findings demonstrate the negative impact of extramedullary disease on outcomes. Other factors influencing post–ASCT outcomes in MM patients include age ≥ 65, treatment response ≥ VGPR before ASCT, light chain ratio ≥ 100 before ASCT, post–ASCT CMV infection, ISS stage III at MM diagnosis, and higher conditioning melphalan dosage ≥ 180 mg/m2. This real-world data provides valuable information for outcome prediction beyond clinical trials.