Proteasome Inhibitor or Immunomodulators Which is The Optimal Maintenance For Newly Diagnosed Multiple Myeloma: A 7-Year Real-World Data in China.

Background: According to different patients’ subgroups choose optimal maintenance therapy. Methods: 226 Newly Diagnosed Multiple Myeloma (NDMM) patients in our center were included, the patients’ characteristics, survival, response, subgroup analysis, adverse reactions were compared between the patients with or without maintenance, proteasome inhibitor (PI) or immunomodulators (IMiDs) maintenance. And the survival of different maintenance duration of bortezomib-based regimens was also analyzed. Results: The maintenance therapy not only upgraded more patients’ response (34.3 vs. 13.3%, p= 0.006), but also signicantly prolonged the patients’ PFS (median PFS: 41.1 vs. 10.5 months, p < 0.001) and OS (median OS: not reached vs. 38.6 months, p < 0.001). Compared with IMiDs, the PFS (median PFS: 43.7 vs. 38.5 months, p = 0.034) and OS (median OS: not reached vs. 78.5 months, p = 0.041) can both benet from bortezomib-based maintenance. The patients younger than 65 years old with bortezomib-based maintenance signicantly prolonged the OS (p= 0.032). Patients achieving the only partial response (PR) after induction and consolidation therapy experienced a signicantly longer PFS and OS with bortezomib-based maintenance compared to IMiDs (p= 0.007, 0.002). Besides, the high-risk patients (ISS 2-3, DS 2-3 and RISS 2-3) with bortezomib-based maintenance can benet PFS (p= 0.002, 0.02, 0.06, respectively) and OS (p=0.059, 0.047, 0.044, respectively) compared with IMiDs. The OS was signicantly prolonged in the patients who received ≥ 12 months of bortezomib-based maintenance than those with maintenance < 12 months (p< 0.001), but no difference was observed in OS between the patients who received the 12-24 or ≥ 24months of bortezomib-based maintenance


Background
Multiple myeloma (MM), a clonal plasma cell neoplasm characterized by monoclonal immunoglobulin production, the second most common hematologic malignancies accounts for about 1-2% of all cancers [1]. In the past two decades, despite autologous stem cell transplantation and novel agents have signi cantly improved the outcome of MM patients, however, most patients experience disease progression or relapse inevitably, maintenance treatment has become a necessary means to deepen the depth of response and sustain the deep response, as well as survival improvement [2,3]. The bortezomibbased regimens are widely used in induction therapy [4][5][6]. Yet its experience in maintenance therapy is still limited. Therefore, we summarized the 7-year follow-up data of MM patients in our center, compared the proteasome inhibitor (PI) bortezomib-based regimens with the immunomodulators (IMiDs) thalidomide and lenalidomide in maintenance treatment to clarify their role in the upgrade of response depth, survival improvement, and evaluate its adverse reactions. By subgroup analysis trying to illuminate the most appropriate maintenance therapy approach for corresponding patient subgroups.

Materials And Methods
Patients Newly diagnosed multiple myeloma(NDMM) patients who had achieved at least a partial response (PR) after induction and consolidation therapy of bortezomib based regimens at the First A liated Hospital of Zhejiang University Medical College from May 23, 2013, to December 13, 2018, were evaluated for inclusion in our retrospective study. All patients were followed up for mortality and survival to October 1, 2020. Demographic and disease characteristics and treatment regimens were extracted from electronic medical records under the approval of review boards. Patients were strati ed according to Durie Salmon (DS) stage, International Staging System (ISS) disease stage, and revised-ISS (R-ISS) stage at diagnosis.
Due to lack of FISH data, 20 patients of the maintenance group (including 15 patients in the PI subgroup, 5 patients in the IMiDs subgroup) and 2 patients of the no maintenance group can't be strati ed by RISS staging.
After 3 to 4 cycles of induction therapy, autologous hematopoietic stem cell transplantation (ASCT) was implemented on some patients eligible for transplantation, according to the patient's age, general state, and patient's willingness. Other patients not eligible for transplantation continued to receive 2 to 4 courses of consolidation therapy which are basically as same as the induction regimens.
After 3 to 4 cycles of induction and consolidation therapy with or without ASCT, the majority of patients received maintenance therapy. A large proportion of patients received proteasome inhibitor bortezomibbased regimens such as PD, PCD, PAD, PTD, VRD for maintenance, with a 3-month cycle, during maintenance all patients received bortezomib lasting 5 weeks on day 1, 8, 15, 22. The other part of patients adopted immunomodulators such as T (thalidomide), R (lenalidomide), TD (thalidomide and dexamethasone), RD (lenalidomide and dexamethasone) as maintenance.

E cacy and Safety Evaluation
Outcome measures included the response to treatment, overall survival (OS), and progression-free survival (PFS). The response to treatment was de ned as the International Myeloma Working Group uniform response criteria, including partial response (PR), very good partial response (VGPR), and complete response (CR) [7]. Adverse reaction assessment adopts the National Cancer Institute Common Toxicity Criteria for AEs, version 5.0.

Statistical Analysis
Statistical analysis was conducted by SPSS 21.0 software (SPSS, Chicago, IL). Baseline characteristics were assessed with descriptive statistical analysis: frequency distribution (n, %) presented for categorical variables and compared using the chi-squared test, median (range) are presented for continuous variables and compared using T-test. PFS and OS analyses were performed using the Kaplan-Meier method, and the log-rank test was used to analyze the differences between survival curves. A value of p < 0.05 indicated statistical signi cance, and all tests were two-sided.

Results
Maintenance vs. no maintenance

Patient Characteristics
The clinical data and biological characteristics of 181 patients with maintenance and 45 patients without maintenance were summarized in Table 1. Signi cant differences can be observed in ASCT. The proportion of patients who adopted ASCT was higher in the maintenance group (17.1 vs. 2.3%). Other baseline characteristics were basically balanced.

Survival
The median follow-up duration for all patients was 36.9 3.6-86.0 months the median PFS was 41.1 (95%CI: 34.5-47.7) months for the patients who received maintenance treatments, and 10.5 (95%CI: 8.0-13.1) months for the no maintenance group (p < 0.001) ( Figure 1A). The median OS of the arm adopted maintenance was not reached, the median OS of the patients who without maintenance was 38.6 (95%CI: 27.0-50.2) months, the survival was distinctly prolonged for patients who adopted maintenance (p < 0.001) ( Figure 1B).

Response
During maintenance, the upgrade of response from PR to at least VGPR was more common in the maintenance group ( 34.3 vs. 13.3%, p= 0.006). The best response after maintenance, the PR rate was 17.7%, and at least VGPR rate was 82.3% in patients who received maintenance versus 42.9% and 57.8% in the no maintenance group (p < 0.001).

Patient Characteristics
Baseline characteristics of 181 patients who received maintenance treatments, including 127 with PI and 54 with IMiDs, were presented in Table 2. No signi cant statistical difference was observed between the two groups of variables. The patients who received PI maintenance were slightly younger, with a median age of 62 years old compared to 63 years old in the IMiDs group. The proportion of patients using PD as the induction regimens was slightly higher in the PI group (16.5 vs. 7.4%). The CR rate after induction and consolidation therapy was higher in the PI group (44.9 vs. 29.6%). In the PI group, there are 7 (5.5%) patients with PAD maintenance, 93 (73.2%) patients with PCD maintenance, 21 (16.5%) patients with PD maintenance, and 6 (4.8%) patients with VRD maintenance. In the IMiDs group, there are 5 (9.3%) patients with R maintenance, 28 (51.9%) patients with RD maintenance, 19 (35.2%) with T maintenance, and 2 (3.7%) with TD maintenance.

Age and Creatinine
For patients younger than 65 years, maintenance with PI signi cantly prolonged the OS (p= 0.032), with a 5-year OS of 81.5 versus 66.1%, respectively. No statistical difference was observed in PFS between the two groups, with a 3-year PFS of 56.6 versus 55%. As for patients aged 65 years and older, there was no signi cant difference in neither PFS nor OS between the two groups. For renal function, OS bene t was observed in the PI group in patients with baseline serum creatinine < 2 mg/dl, with a 5-year OS of 76.8 vs. 64.9% (p=0.036). In those patients with normal renal function, PI maintenance therapy also improved the PFS compared with the IMiDs group, with a 3-year PFS of 58.4 vs. 52.0% (p= 0.07), although the statistical difference was not signi cant. In the patients whose baseline serum creatinine ≥ 2 mg/dl, there was no difference in PFS and OS between the two maintenance treatment options (Table 3).

Response after induction and consolidation therapy
Patients achieving the only PR after induction and consolidation therapy experienced a signi cantly longer PFS and OS with PI maintenance compared with IMiDs, with 3-year-PFS of 44.8 and 25.0% (p= 0.007), 5-year-OS of 77.9 and 45.1% (p= 0.002). However, in the patients achieving at least VGPR, no difference was observed between the two groups in PFS or OS (Table 3).

Clinical Stages
Patients were strati ed according to clinical stages. Patients in ISS 1, DS 1, and RISS 1 were classi ed as non-high risk; the other patients in ISS 2-3, DS 2-3, and RISS 2-3 were classi ed as high risk. Overall, highrisk patients receiving PI maintenance therapy can improve survival.

Adverse Reactions
During maintenance treatment, the incidence of second primary malignancies was higher in the IMiDs group (0 vs. 3.7%, p= 0.029). There were no signi cant statistical differences in other adverse reactions between the two maintenance treatment options. The most common hematological adverse events are thrombocytopenia (5.5 vs. 7.5%) and Neutropenia (4.7 vs. 3.7%). For non-hematological adverse events, the most frequent were peripheral neuropathy (23.0 vs. 33.3%) and infection (22.0 vs. 25.9%). The results were presented in Table 4.

Treatment Duration of Proteasome Inhibitor Maintenance
Proteasome Inhibitor bortezomib-based regimens are the main maintenance therapy options in recent years, with 3-month a cycle. This paradigm of long-term treatment needs to consider many other factors such as patients' quality of life, convenience, and burden of long-term treatment. Thus, we perform a secondary analysis to nd the optimal treatment duration. The median treatment duration of the patients who received bortezomib-based maintenance after induction and consolidation therapy was 12.9 (0.8-45.1) months. The PFS was improved with the increasing treatment duration (p< 0.001). The OS was signi cantly prolonged in the patients who received ≥12 months of bortezomib-based maintenance than those with maintenance <12 months, with the 5-year OS of 91.9 vs. 51.1% (p< 0.001) ( Figure 3A). However, no difference was observed in OS between the patients who received the 12-24 or ≥ 24months of bortezomib-based maintenance (p= 0.292), with a 5-year OS of 96.4 vs 86.6% ( Figure 3B).
The relationship between the maintenance treatment course and the outcome is basically consistent with the above result. The median maintenance cycles were 4 (1-14) of the patients completed after induction and consolidation therapy. The PFS was improved with the increase of the number of the treatment course (p< 0.001), the OS was signi cantly different in patients with < 4 vs. ≥4 maintenance treatment courses (p< 0.001), with a 5-year OS of 42.3 vs. 92.7%. The OS between patients with 4-9 vs. ≥ 9 maintenance treatment courses was not statistically signi cant (p=0.214), with a 5-year OS of 95.5 vs. 83.7%.

Discussion
In the past two decades, with the introduction of new drugs such as proteasome inhibitor bortezomib, and the immunomodulators thalidomide and lenalidomide, the treatment paradigm of NDMM has changed dramatically. Those new drugs have less toxicity than traditional chemotherapy drugs, allows long-term maintenance becoming a treatment paradigm. Maintenance therapy usually refers to administering a course of long-term low-dose chemotherapy after induction and consolidation therapy. In our study, the signi cance of maintenance treatment not only improved more patients' response (34.3 vs. 13.3%) but also signi cantly prolonged the patients' PFS (median PFS: 41.1 vs. 10.5 months) and OS (median OS: not reached vs. 38.6 months). We admit that the PFS was shorter compared with the literature in no maintenance group [8][9][10]. The reasons are the following: rstly, almost all of our no maintenance patients did not undergo ASCT, moreover, the patient did not receive maintenance partly because their poor physical condition cannot tolerate maintenance. Besides, our limited sample size may impact results.
Various maintenance treatment options in NDMM are recommended within consensus and guidelines [15][16][17]. As the consensus of Mayo Clinic recommends: immunomodulators maintenance is recommended for patients with standard-risk cytogenetics, meanwhile, PI maintenance for high-risk cytogenetics [16]. Data from our center revealed that compared with thalidomide and lenalidomide, the PFS and OS can both bene t from bortezomib-based regimens (median PFS: 43.7 vs. 38.5 months; median OS: not reached vs. 78.5 months), which was consistent with the literature [18]. It is worth noting that in the phase III HOVON-65/GMMG-HD4 trial the early research report showed the OS was superior in bortezomib-based regimens induction and maintenance arm to non-bortezomib-based regimens induction and thalidomide maintenance arm [19], however, after long-term follow-up, the OS was no difference between two arms, due to the majority of patients have relapsed inevitably and received multiple effective post-relapse treatments [20].
To further explore the best maintenance options for patients in different subgroups, we conducted a more detailed subgroup analysis. In younger patients, the OS was more superior in patients with bortezomibbased maintenance (median OS: not reached vs. 78.5 months). Bortezomib does not need dose adjustment for MM patients with renal impairment [21]. Moreover, bortezomib-based regimens before and after autologous stem cell transplantation (ASCT) can overcome the adverse effects of renal damage on prognosis [19,20,22]. Our data showed that patients with baseline serum creatinine < 2mg/dl, both PFS and OS can bene t from bortezomib-based maintenance, whereas, the patients with baseline serum creatinine ≥ 2 mg/dl, no statistical difference was observed between the two maintenance options. It may because most of our patients did not undergo ASCT, the bortezomib-based regimens induction and maintenance treatment without ASCT is not enough to overcome the adverse effects of renal impairment on prognosis. Moreover, due to our limited sample size, we need to interpret this conclusion carefully and expand the number of patients to verify the conclusion. The response after induction and consolidation was also an important factor to choose maintenance treatment options, in the patients who achieved only PR after induction and consolidation, the bortezomib-based maintenance signi cant prolonged the PFS and OS, however, in the patients who achieved at least VGPR, there was no statistical difference in PFS and OS between two groups. Clinal trial research showed that bortezomib consolidation after ASCT only improved the PFS of patients not achieving at least VGPR, no effect on patients who achieved at least VGPR, and did not prolong OS in both category patients [23]. Consolidation therapy refers to utilize a short course of treatment for reducing the number of residual tumor cells, which cloud prolonged the PFS but not OS[18]. Consolidation and maintenance play a different role in patients' outcomes. In high-risk patients (including ISS 2-3, DS 2-3, and RISS 2-3), it's very necessary to choose bortezomib-based maintenance, the PFS and OS were signi cantly superior to thalidomide or lenalidomide maintenance. Yet, in the non-high-risk patient, the difference did not reach statistical signi cance between the two groups.
The NCCN guidelines have added bortezomib-based maintenance as an option for patients with or without ASCT [24]. In PI maintenance, the PCD regimen was the most adopted option, due to its greater e cacy, fewer adverse reactions, lower cost, and convenience to use. It is the rst-line treatment for NDMM approved by multiple centers in induction therapy [25][26][27], which is also be widely used in maintenance. The treatment duration usually lasts 2-3 years or until the disease progresses. We performed a secondary analysis to found the optimal treatment duration, the optimal duration that the patient can not only achieve the best disease relief but also reduce the burden of long-term maintenance on the patients. Through our analysis, the bortezomib-based maintenance duration lasting 12-24 months after induction and consolidation therapy can reach a satisfactory OS, the outcome of maintaining more than 24 months is similar to that of 12-24 months. It is equivalent to 4-8 courses of maintenance treatment after induction and consolidation treatment, with 3-month a cycle.
The rst oral bioavailable PI ixazomib was licensed for the treatment of MM in 2018 in China. Multiple studies have con rmed its effectiveness and safety [28,29]. Besides, ixazomib has become a category 1 "other recommended" maintenance option for patients eligible for transplantation in NCCN guidelines [24]. Ixazomib may play an important role in the future maintenance paradigm. Various studies are ongoing to explore the best maintenance according to different patient subgroups and optimize individual patient outcomes, at the same time, may be able to provide more different maintenance options. In the future maintenance paradigm not only e cacy and adverse effects be considered, but also the quality of patient life, convenience, compliance of long-term treatment, and nancial burden. In addition, the role of minimal residual disease (MRD) in guiding maintenance treatment decisions has been increasing attention. The maintenance paradigm will become increasingly well-established.

Conclusion
In our study demonstrate the e cacy of bortezomib-based regimens as maintenance. Besides, we have some suggestions of optimal maintenance therapy according to different patient subgroups. And we provide a reference for the optimal maintenance duration in the real world. Due to sample size limitations, we may need to interpret a few conclusions carefully. At the same time, we will continue focusing on maintenance therapy in the future to perform research with larger sample size, longer follow-up, and more maintenance options.

Declarations Ethical Approval
The study was approved by the Ethics Committee of the First A liated Hospital of Zhejiang University School of Medicine, with the ethical number of (2020) IIT (822). We got the patient's informed consent and promised that all patients' information was strictly con dential.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare no competing interests