Autologous Stem Cell Transplantation in Multiple Myeloma with Renal Failure: Friend or Foe?

Autologous stem cell transplantation (ASCT) is a standard treatment for multiple myeloma (MM), but the clinical response and renal curative effect in MM patients with renal failure (RF) remain controversial. The myeloma kidney disease has different types, and most are due to the direct toxic effects of light chain. Although ASCT can effectively clear the light chain, the data of renal function improvement are still limited. We reviewed the published literatures, focusing on the prospective studies, the retrospective analysis studies, and the case reports. RF patients who received ASCT displayed a low survival rate (OS: HR 1.95, 95% CI 1.020 to 3.720; I2 = 64.9%, P = 0.014) and a shorter EFS/PFS (EFS/PFS: HR 1.53, 95% CI 1.090 to 2.140; I2 = 0%, P = 0.669). However, ASCT was feasible and could have the similar clinical response outcomes compared with the normal renal function (CR: OR 1.013, 95% CI 0.569 to 1.804; I2 = 48.5%, P = 0.101; PR: OR 1.013, 95% CI 0.342 to 1.226; I2 = 46.3%, P = 0.144). Moreover, MM with RF after ASCT had a good improvement of renal function and melphalan is still an important factor affecting the treatment of ASCT.


Introduction
Renal failure (RF) is one of the most common complications of multiple myeloma (MM), and it has been associated with higher risk of mortality and increased hospitalization rates due to complications such as electrolyte abnormalities, catheter-related complications, and infections [1,2]. Factors contributing to myeloma kidney disease include hypercalcemia, dehydration, hyperuricemia, amyloid deposition and plasma cell infiltration, light chain-induced proximal tubular damage, cast nephropathy, and interstitial nephritis [3]. Furthermore, administration of nephrotoxic medication, dehydration, and hypercalcemia always adds to the development of acute kidney injury (AKI) [4][5][6]. Improved renal function is an important therapeutic aim and has become a predictor of better outcome in MM [7].
Autologous stem cell transplantation (ASCT) as a standard treatment for MM because of its association with longer event-free survival (EFS) and higher complete response (CR) rate [8,9], it has been the mainstay of therapy in young patients (age < 65 years) with MM [10]. Historically, MM with RF appeared to have higher rates of transplant-related mortality (TRM) compared with the normal renal function (NRF) patients [11]. Although ASCT is still one of the disease's most effective treatments [12], the presence of coexistent renal disease limits the therapeutic options and stem cell transplant eligibility [13]. In recent years, several reports have shown that the use of ASCT is safe and effective in MM with RF [14][15][16][17][18]. However, there still have some considerable variabilities in reported survival outcomes and renal recovery from the limited literature, and the studies included have different priorities in clinical and renal response.
Herein, we fully summarized the studies of ASCT in MM with RF, including the prospective studies, the retrospective studies, and the case reports. The diagnosis, types, and mechanisms of RF in MM are also discussed. More importantly, we analyzed the data of renal recovery and clinical response to answer the question of clinical controversy following ASCT treatment and evaluate whether MM with RF benefits from ASCT or not.

Diagnosis, Types, and Mechanisms of RF in MM
2.1. Diagnosis of RF in MM before the ASCT Therapy. The classification guidelines for renal failure in MM were adapted in 2014 [19]; eGFR was used only in patients with stable renal function. From the studies we included, most of them were according to the novel International Myeloma Working Group (IMWG) criteria for symptomatic MM [19], and it is based on either reduced creatinine clearance (CrCl < 40 ml/min) or elevated serum creatinine (SCr > 2 mg/dl). Although the criteria are more sensitive for the determination and evaluation of renal failure in nephropathy, the standards of RF in our included studies are still inconsistent; the diagnosis of renal failure in MM requires relatively uniform standards in the future.  [20,21]. Moreover, nephrotoxic drugs (aminoglycoside antibiotics and nonsteroidal antiinflammatory agents), hypercalcemia, dehydration, and contrast agents contribute to the development of renal failure [5,22,23]. As a result, the excess LCs form casts and aggregates with uromodulin in the distal nephron, leading to tubular obstruction and concomitant inflammation [20,21,24]. Furthermore, LC has direct toxic effects on kidney damage, and LC protein accumulates in renal tubular epithelial cells, inhibiting the metabolism of tubular cells and affecting the transportation of normal ions, amino acids, phosphates, etc. With the cast nephropathy developed, LCs can infiltrate the whole kidney and cause tubular, vascular, or glomerular damage. ASCT can effectively clear the LC, and renal damage may achieve remission; however, the data of renal function improvement are still limited.

ASCT in MM with RF: The Summarized Clinical Studies
There is growing concern about the curative effect of ASCT in MM with RF; more studies were reported to assess the clinical response and renal function in recent years. We fully summarized those studies but the included studies have different types, and the data of those studies were incomplete and variable. Therefore, we classified these studies into the cohort studies, the retrospective analysis studies, and the case reports, and the characteristics of each study are shown in Table 1. We fully summarized and classified the data of RF diagnosis, conditioning regimen, clinical response, survival, and response of renal function. Furthermore, we discovered that the present studies have different priorities in clinical and renal response; in the cohort studies, authors seemed to attach more weight to the clinical response. On the contrary, a retrospective analysis took more attention to renal function change. We also took a meta-analysis through the cohort study data to discuss whether the use of ASCT is safe and effective in MM with RF or not, and the data included the survival analysis, clinical response, and mortality. The following search terms were used: (1) "Autologous stem cell transplantation" or "Monoclonal Gammopathies" or "ASCT"; (2) "renal failure" or "renal function" or "acute kidney injury"; (3) "multiple myeloma" or "myeloma" or "MM"; and (4) "the cohort studies," "the retrospective analysis studies," and "the case report studies." In addition, the reference lists of retrieved papers and recent reviews were reviewed. The flow diagram of search strategy is presented in Figure 1.

Study Criteria.
The inclusion criteria for studies were as follows: (1) the cohort studies comparing data on the clinical response and survival ("CR," "PR," "VGPR," "OS," "EFS," "PFS," and "TRM"); (2) validated diagnosis of renal failure and original research related to renal failure in MM patients; (3) studies that provided information about ASCT in MM with renal failure; and (4) articles that reported a clear comparison of RF (renal failure) population versus NRF (normal renal function) population controls with a direct effect on the clinical response and survival data.
The exclusion criteria were as follows: (1) duplicate studies; (2) studies such as systemic reviews, meta-analyses, and comments; and (3) studies of ASCT in MM with renal failure without detail research data in the clinical response and survival data.

Data Extraction.
Data extracted from each study included the first author's name, the publication year, the country of study origin, number of patients, median age, and the clinical response and survival ("CR," "PR," "VGPR," "OS," "EFS," "PFS," and "TRM"). If a study did not clearly mention any of the above key points, it had not performed the required methods. Two of the authors (Hongfei Zhong and Gaosi Xu) independently reviewed the selected studies and extracted data. Discrepancies were resolved by discussion.
3.1.4. Statistical Analysis. The data was abstracted and analyzed using Stata (version 12) to make the outcomes more comprehensive. The binary variable outcomes were the EFS/PFE and OS; the data were expressed as the hazard ratio (HR) with 95% CI (confidence interval), and the estimation  Off topic (n = 165) Not cohort or retrospective analysis studies or case report (n = 38) System review or meta-analysis (n = 4)

Studies included in qualitative synthesis (n = 20)
The cohort studies (n = 6) The retrospective analysis studies (n = 9) The case report studies (n = 5) Studies included in qualitative synthesis (meta-analysis) (n = 6) Figure 1: Flow diagram representing the selection process.
of the effect was performed by using a random effects model. Other binary variable outcomes were the PR and CR, and the date were expressed as the odds ratio (OR) with 95% CI (confidence interval); when combining studies, the random effects model was used to account for study heterogeneity. We used Q statistic and I 2 tests to evaluate the heterogeneity. Low, moderate, and high heterogeneities were represented by thresholds of <25%, 25-75%, and >75%, respectively. P ≤ 0:05 was considered significant in all statistical tests.

Data Analysis.
Recently, some studies reported the safety and clinical efficacy of ASCT use in myeloma patients with RF (Table 2) [16,[25][26][27][28][29]. Six articles [16,[25][26][27][28][29] with a total of 2930 MM patients were included in the meta-analysis. The binary variable outcomes were the incidence of overall survival (OS), event-free survival (EFS), progression-free survival (PFS), complete response (CR), partial response (PR), very good partial response (VGPR), and transplantation-related mortality (TRM). In addition, the data of OS and EFS expressed as the hazard ratio (HR) with 95% confidence interval (CI), and the data of CR, PR, VGPR, and TRM were expressed as the odds ratio (OR) with 95% CI; the estimation of the effect was performed by using a random effects model. The clinical response and survival analysis in MM with RF after ASCT are shown in Figure 2.
To the best of our knowledge, this is the only and the first meta-analysis that reported the clinical response and survival data of ASCT treatment in MM with RF. Obviously, the results showed that the use of ASCT was associated with increased risk of mortality, and the outcome was consistent with the previous studies [13]. The CR (OR 1.013, 95% CI 0.569 to 1.804; I 2 = 48:5%, P = 0:101) and PR (OR 1.013, 95% CI 0.342 to 1.226; I 2 = 46:3%, P = 0:144) were not significantly different between the RF and NRF groups. Survival analysis indicated that MM with RF have lower survival rates (OS: HR 1.95, 95% CI 1.020 to 3.720; I 2 = 64:9%, P = 0:014), and the major cause of a low survival rate in MM with RF may be due to the high toxicity in ASCT therapy. As a whole, ASCT was feasible and could lead to similar clinical response outcomes compared with those without advanced renal failure, but the survival analysis seemed to be not optimistic. Moreover, we noticed that the number of patients in some studies was relatively small. So large-size cohort studies are needed to prove this conclusion of ASCT for MM with RF in the future. Unfortunately, these reports had limit outcomes of renal response; only three studies [16,26,29] reported the renal function change.

ASCT in MM with RF: The Retrospective Analysis Studies.
Nine retrospective analysis studies reported the outcome of ASCT treatment in MM with RF, these studies were done to mainly observe the alteration of the RF in MM patients. It was revealed that few studies focus on the clinical response and survival data and most retrospective studies tend to observe the renal response, and it was contrary to the emphasis of previous cohort studies [16,[25][26][27][28][29]. In general, fewer clinical response (CR, PR, and VGPR) was reported in the retrospective analysis studies. From the existing data, ASCT treatment seemed to have a better PR rate (62%), and the CR was 38% (  [18,32]. We indicated that the dose of melphalan escalation may result in higher response rates. Overall, from the retrospective studies, we conclude that ASCT as a good clinical response treatment could be an effective therapy in MM with RF. Although cohort studies [16,[25][26][27][28][29] reported the clinical efficacy of ASCT use in MM with RF, however, the data of renal function response was less. We summarized the retrospective analysis studies that reported renal function response, and these studies complemented the renal response outcome of the previous cohort studies. Parikh [36], and Tosi et al. [31] reported the renal response after ASCT, and the improvements in renal function were 32%, 25%, 60%, 17%, 33%, 100%, and 83%, respectively. However, the definition of RF in each study was different. From the limited research, we found that lower-dose melphalan might have a better improvement of renal function (Augeul-Meunier et al. 60%, Glavey et al. 100%, and Tosi et al. 83%, respectively). On the contrary, the improvements of patients with renal recovery in the high-dose melphalan group were 32% and 25%. What is more, the USA Myeloma Group reported that the patients with RF underwent ASCT and ten patients (21%) experienced downstaging of renal failure [33]. It also reminds us that high doses of melphalan are associated with severe renal failure and should be used cautiously. On the other hand, age may also be an important factor affecting the curative effect of ASCT treatment. Tosi et al. [31,37] reported a good renal function improvement, and the median ages were 49 and 47. A previous study also indicated that ASCT has been the mainstay of therapy in young patients with MM [10]. ASCT treatment may have age limitations, especially in patients with RF. However, some researches associated with older patients still have a safe and efficacy treatment of renal recovery [30,36]; controversies exist about the benefits of transplantation for patients with older age. A future study needs to assess the effects of age values at the time of ASCT treatment in MM. Furthermore, patients in four retrospective analysis studies suffered a predialysis before ASCT [30,33,34,36]. It appears from the data at hand that there is almost no connection between the predialysis and the outcome of ASCT therapy.

ASCT in MM with RF:
The Case Report Studies. Five case report studies [37][38][39][40][41] were included in our research, and our summary is shown in Table 4. One patient reported an acute renal tubular necrosis, which may due to the consumption of cooked grass carp fish in the night. In contrast to those of other patients in the four studies, the renal functions were improved.
Two studies (Bigé et al. and Tauro et al.) have shown a renal improved advantage for patients who receive ASCT with a high-dose melphalan (200 mg/m 2 ) treatment; this is in contrast to our retrospective study data. Historically, patients with RF either have received reduced doses or have been excluded from ASCT therapy with high-dose melphalan. Perhaps, the researchers prefer to report that high-dose melphalan may be safely administered to MM with RF. However, cohort studies with more patients are still necessary to assess the benefit of high-dose therapies in these cases.

Melphalan: Is It Safe for MM with RF?
Melphalan is probably the most effective chemotherapeutic agent in MM with a clear dose-response effect, and melphalan usually is a conditioning regimen before ASCT treatment. It has shown reduced overall mortality and improved PFS compared to conventional chemotherapy in MM [8,9,41,42]. The standard conditioning regimen of melphalan (a dose of 200 mg/m 2 ) was used for patients with NRF [43], melphalan has a dose-response antimyeloma effect, and higher doses could potentially improve the clinical response when used as a conditioning regimen for ASCT [44]. Unfortunately, melphalan has encountered dose-limiting toxicities, especially in MM with RF. Because of conflicting data on altered melphalan pharmacokinetics in renal insufficiency, patients with creatinine levels > 2 mg/dl have usually been excluded from high-dose melphalan treatment [45,46]. However, some studies have found high-dose chemotherapy with melphalan can be administered to selected patients with RF [34,40]. Our two case reports also come to the same conclusion [39,41], and RF might no longer constitute a criterion for dose reduction or exclusion from such therapy.
In our summarized clinical studies, the data associated with melphalan dose were chaotic, and most studies showed that the dose of melphalan use was arbitrary (from 100 to 200 mg/m 2 ), and the definition of high-dose melphalan was different in each study [30,33]. In the cohort study groups, five researchers reported the use of melphalan as the conditioning regimen during the ASCT treatment [16,[25][26][27][28]; the dose of melphalan use may be the source of heterogeneity in the meta-analysis. Owing to the limited data of dose gradient of melphalan use, we cannot take a subgroup to assess whether the dose gradient of melphalan will affect the survival analysis of ASCT treatment in MM with RF. However, existing data concluded that remission rate may not be affected by the melphalan use (CR: OR 1.013, 95% CI 0.569 to 1.804; I 2 = 48:5%, P = 0:101; PR: OR 1.013, 95% CI 0.342 to 1.226; I 2 = 46:3%, P = 0:144), and the heterogeneity of data was acceptable. In the retrospective analysis studies, six studies used melphalan as the conditioning regimen, and we indicated that low-dose melphalan (melphalan 80 mg/m 2 , 140 mg/m 2 ) treatment might have a lower mortality [30,32], but with the increase of melphalan doses, the TRM was increased [18]. Furthermore, low doses of melphalan use may achieve a good PR [30,31], and high doses might have a good benefit in CR [18,32]; the dose of melphalan escalation may result in higher response rates. We also found low-dose melphalan (melphalan 80 and 140 mg/m 2 ) treatment might have a lower mortality [30,32], but with increasing doses of melphalan, the data of survival analysis was controversial. As for the renal improvement aspect, low-     dose melphalan use has demonstrated a good renal recovery from the retrospective studies. However, with the process of the increased dose, changes in renal function have been described in different outcomes, so clinical trials are required for more evaluation of high-dose melphalan use in MM with RF, especially in renal recovery outcomes.

Conclusion
Accumulating evidence suggests that in MM with RF, ASCT could be a feasible therapy and can lead to similar remission outcomes to those without advanced RF. Our current study indicated that the MM with RF after ASCT truly has a good improvement of renal function but has a low survival rate.
For the recovery of kidney function in MM patients, ASCT may probably be a friend, but it may be a foe due to the low survival rate. In general, from the overall efficacy, ASCT is worth a try in MM patients with RF. The clinical response of the conditioning melphalan therapy in RF patients remains controversial, especially in dose response of melphalan use. Moreover, melphalan is still an important factor affecting the treatment of ASCT. Overall survival EFS: Event-free survival PFS: Progression-free survival CR: Complete response PR: Partial response VGPR: Very good partial response TRM: Transplantation-related mortality.

Additional Points
Highlights. (i) The studies of ASCT in MM with RF were fully summarized. (ii) The diagnosis, types, and mechanisms of RF in MM were discussed. (iii) The data of renal recovery and clinical response during ASCT treatment were analyzed. (iv) Whether MM with RF benefits from ASCT or not was evaluated. (v) Melphalan is still an important factor affecting the treatment of ASCT.

Conflicts of Interest
No any conflict of interest was reported in this work.