ALC reflects the restoration of hematological parameters after autologous peripheral blood (PB) stem cell transplantation, and is an independent prognostic factor for clinical outcome. A retrospective study conducted in 537 newly diagnosed MM patients in the Mayo Clinic indicated that the survival of MM patients with an ALC > 1.4 × 10⁹/L was associated with better overall survival (OS) (65 vs. 26 months) (2). Another study revealed the same results, in which, after induction therapy day (D) 29, 38 MM patients with an ALC > 0.8 × 10⁹/L had better OS (58.3 vs. 42.5 months) (3). Patients with ALC ≥ 1400 cells/µL or <1400 cells/µL at post-autologous stem cell transplant at D0, D15, and D90 experienced a different OS (111, 90.7, and 84 months vs. 74, 70.5, and 65 months, respectively) (4). In D23 of post-autologous stem cell transplant, MM patients with ALC ≥ 1000/mm³ vs. < 1000/mm³ also showed a different OS (37.96 vs. 23.19 months) (5). However, ALC in 125 older MM patients treated at diagnosis with IMIDs and not eligible for autologous stem cell transplantation (ASCT) is unrelated to OS (6). ALC could reflect host immune function. The survival of MM cells mostly affected by their interaction with the immune micro environment. ALC can be a most reported index to predict the survival of MM patients. However, there some question need to be solved carefully. 1) what is the time we should detect ALC, new diagnosis point or post-autologous stem cell transplant at some day? 2). As we all known that the recovery of immunological cell after hematopoietic cell transplantation need almost 1 year (34),current data need extend the follow up time to further observe the survival predictive value of ALC in MM.

The NLR has been reported as an adverse prognostic factor among cancer patients (35–37). Lately, NLR has been reported to possess prognostic value in patients with MM: high NLR is associated with the score of the ISS system in newly diagnosed MM patients. Furthermore, MM patients have shortened OS with high NLR (7–12, 14, 38).

Engin Kelkitli et al. (39) first reported that, in 151 MM patients, the NLR was significantly higher than that in healthy controls (2.79 ± 1.82 vs. 1.9 ± 0.61). However, because the patients included in these studies were not in full accord, we did not obtain an accurate and concrete NLR value. Romano et al. (8) showed that the median NLR in 309 newly diagnosed MM patients was 1.9 (range: 0.4–15.9). This was similar to another study that evaluated the median NLR in 131 MM patients and the value was estimated to be 1.93 (range: 0.10–36.23) (9). In a recent study, 559 MM patients were included. The NLRs of patients prior to therapy and 123 healthy controls were 2.096 ± 0.0629 and 1.771 ± 0.0747, respectively (10). As NLR data were obtained by estimating complete blood count (CBC), differences in NLR might result from the different numbers of MM patients considered in the various studies; however, the results indicated an increase in NLR in MM patients.

The NLR at diagnosis can help predict the survival of patients with MM. Several studies have revealed that a high NLR is associated with shorter OS or PFS. In several studies (7, 9, 11), multivariate analysis for OS showed that high NLR was an independent significant prognostic factor. Shi et al. (10) performed a multivariate analysis, which indicated that elevated NLR was a statistically independent predictor of PFS. Engin Kelkitli et al. revealed that the duration of OS and EFS in MM patients with NLR ≥ 2 was shorter than that in patients with NLR < 2 (5-year OS: 87.5% vs. 42.4%; 5-year EFS: 88.4% vs. 41.8%) (7). In a similar study, the 5-year PFS and OS were 18.2% and 36.4% in 309 newly diagnosed MM patients with NLR ≥ 2 compared to 25.5% and 66.6% in patients with NLR < 2 (8). In another report, 52 MM patients were enrolled, and patients with NLR ≤ 1.72 at diagnosis obtained superior OS rates than MM patients with NLR > 1.72 (42.75 vs. 26.14 months) (12). In a study involving 559 MM patients, Shi et al. reported that the median PFS and OS were significantly shorter among MM patients with NLR > 4 compared to the rest of the cohort (PFS: 24.03 vs. 37.46 months; OS: 43.2 vs. 56.0 months) (10).

As for the value of NLR in the treatment of MM patients, there have been a few studies conducted on the application of drugs and ASCT. The application of novel drugs, such as proteasome inhibitors and immunomodulatory agents, is a remarkable improvement in MM treatment. As a result, the OS of MM patients has shown recent improvements (2). Zhou, X. in 2018 (13) investigated 76 newly diagnosed MM patients to study the association between NLR and OS. In multivariate analysis, NLR was not an independent prognostic factor for poor survival of patients receiving bortezomib-based therapy; patients obtained lower 4-year OS rates with NLR > 2.95 compared to patients with NLR < 2.95 (30.9% vs. 64.8%). However, in a prior study involving 179 MM patients treated with the bortezomib-melphalan-prednisone (VMP) regimen, multivariate analysis revealed that high NLR was an independent poor prognostic factor (14). In addition to the OS, it also has an indication of the treatment response to bortezomib. The complete response rate (CRR) was significantly inferior when comparing high- and low-NLR groups (7% vs. 26.1%). Based on the results of univariate logistic regression analysis, as well as multivariate analysis, high NLR was identified as a poor prognostic factor of CRR to bortezomib (14). In the last decade, ASCT has been widely used in treatment of MM patients. An increased NLR indicates inferior survival of MM patients with ASCT (8, 17). Romano, A. et al. showed that the median PFS was significantly shorter for MM patients with NLR ≥ 2 compared to those with NLR < 2 (22.1 vs. 43.4 months) at diagnosis, when considering the clinical outcomes of ASCT (8). In 2018, Solmaz Medeni, S. (17) conducted a study involving MM patients who underwent ASCT. High NLR at the 100^th day post-transplantation is associated with inferior OS and PFS.

In summary, for newly diagnosed or post-ASCT MM patients, increased NLR may help predict inferior clinical outcome, and it may be used as a prognostic biomarker for the prediction of survival and treatment response owing to the low cost and rapidity of the test. However, the time to detect NLR and the cutoff value should be further standardized before it can be used in clinical practice.

The predictive role of PLR in MM patients remains controversial. A previous study revealed that PLR could be used as an independent prognostic factor in several cancers. A retrospective study that enrolled 175 MM patients revealed that the median PLR was 127.69 (range: 0.46–1959.60). Furthermore, the best cutoff value of PLR for OS by ROC curve plot was 155.58. MM patients with a PLR> 155.59 also showed lower albumin levels and higher survival staging (9). A high PLR on the 100^th day post-transplantation indicated an inferior clinical outcome in 150 MM patients after ASCT (39). However, in another study that enrolled 315 newly diagnosed MM patients, there were no significant differences in OS or PFS, in contrast to PLR. In this study, NLR might be used as a superior index to predict survival than PLR. Based on the results of multivariate Cox analysis, it can be inferred that PLR is not a useful independent prognostic factor and it cannot be used to predict the OS and PFS of MM patients (15). All those results may because PLR takes both promote tumor status and anti-tumor immune status into consideration,which lead this index become not exactly since there need take platelet and lymphocyte into consideration at the same time.

MLR is another useful index for predicting the survival of individuals upon diagnosis (40). In a previous study that enrolled 285 MM patients (16) at a diagnosis phase, patients with MLR ≥ 0.24 had shorter OS and PFS than patients with MLR < 0.24. Additionally, multivariate analysis revealed that MLR ≥ 0.24 could be used as an independent predictor for OS and PFS. In another study that enrolled 150 MM patients after ASCT, at the 100^th day of a post-transplantation period, MLR ≥ 0.27 indicated an inferior clinical outcome (17).

Numerical and functional T cell abnormalities have been described in MM patients, which results in the occurrence of immune dysfunction, such as disrupted immune surveillance and immune escape, and can even lead to disease progression (41, 42). Abnormal T cell repertoire owing to an abnormal ratio between CD4⁺ and CD8⁺ T cells, a decrease in the number of CD4⁺ T cells, and an increase in the number of regulatory T cells (Tregs) have been reported in previous studies in MM patients (41, 43, 44). As for functional dysfunction, the abnormally high expression of immune checkpoints, such as programmed cell death ligand 1 (PD-L1), leads to the inhibition of CTL cells in MM (45).

The number of CD4⁺ and CD8⁺ T cells plays a role in predicting the survival of MM patients. Schmidmaier R. et al. used flow cytometry to detect PB lymphocytes and aphaeresis products (AP) in 41 MM patients. Increased number of CD4⁺ cells and increased ratio of CD4/CD8 are significantly correlated with prolonged EFS. However, a high proportion of HLA-DR positive lymphocytes is negatively associated with EFS and OS (46).

The amplified T cell clones improve the survival of MM patients, and T cell expansion is predominantly CD8⁺ (93%). The expansions are associated with a significant prolongation of PFS and OS. This finding was similar to that observed in the thalidomide arm amplification group (47). In a clinical trial, 75 patients with relapsed/refractory myeloma received thalidomide. Elevated vascular endothelial growth factor (VEGF) baseline values help predict excellent RR and PFS. The increased number of CD57⁺ cells indicates a better PFS (18). Another subgroup of T cells also play a role in predicting survival. A study of 85 autologous HPCT patients (including 11 with MM) revealed that the number of CD4⁺ T cells pre- transplantation was associated with PFS and OS in patients with hematologic malignancies. The number of pre- transplantation memory T cells (CD4⁺CD45RA^-CD62L^-) was associated with PFS (48).

However, there need more studies focus on immune checkpoints such as CTLA4, TIGIT, and VISTA, and their roles during the progression of myeloma. Those results can reveal the role of immune cell in MM patients survival prediction.

A previous study revealed that OS was 2.8 years for low counts of CD19⁺ B cells (<125 cells/mL), whereas the OS for the high-CD19⁺ B cell count group (> 125 cells/mL) was 4.0 years in newly diagnosed MM patients (19). Based on the data obtained for 101 consecutive MM patients, the increased total PB counts of CD19⁺ B cells at pre-AHSCT is significantly associated with improved 2-year PFS (83% vs. 53%) and OS (93% vs. 63%). The same result was observed in bone marrow samples of MRD-positive patients as determined by flow cytometry pre-AHSCT. However, there was no association observed between pre-AHSCT total PB CD19+ B cell count and PFS or OS for MRD-negative patients. This indicated that a high B cell count improved the therapy outcomes. Higher counts of sub populations of B cells, including naive and memory B cells, in PB are associated with a superior survival and improved values of 2-year PFS and OS (20). CD19⁺ B cell counts in the PB and bone marrow are significantly higher in long-term disease control MM patients than those in healthy donors or MGUS (49).

Not only B cells, but also the B cell subgroup are associated with OS in MM patients. A real word data from CHINA analysis bone marrow Bregs in 29 MM patients. Patients with < 10% (0-62.1%) Bregs (within CD19⁺B-cell compartment) had significantly worse OS (21) with levels of serum B cell maturation antigen (sBCMA) are associated with PFS and OS in MM patients; sBCMA > 326.4 ng/mL had inferior PFS and OS (22).

All the data indicated that B cell (including B cell sub population) may play a vital role in the development of MM. Since plasma cell come from B cell. More B cell population may inhibit the proliferation of malignant plasma cell directly or indirectly, another possible reason may the decrease of malignant plasma population make the population of B cell or normal plasma cell increased. It still need further study to reveal the underlying mechanism of how B cell regulate the proliferation of MM cell. But at least, our present data can exactly indicate that B cell can well predict the prognosis of MM patients.

Tregs are a sub population of T cells that control autoimmune reactivity in vivo and can suppress immune responses by directly interacting with other immune cell types or by the secretion of immuno suppressive cytokines. It is divided into naturally regulated T cells (nTregs) and induced or adaptive regulatory T cells (aTregs or iTregs).

In MM patients, Tregs showed no significant association with major clinical and laboratory characteristics after a median follow-up of 33 months. From a functional perspective, Tregs exhibit potent inhibitory effects regardless of the disease status (50). In the CD4⁺ T cell subset, the balance between Th17 cells and immuno suppressive Tregs is an important factor in immune control of malignant tumors. There seems to be a clinical significance between Th17 and Tregs (51). A study conducted by Bryant and his colleagues found that long-term survival of MM patients (>10 years after diagnosis) was significantly associated with higher Th17/Treg ratio than patients who were subjected to follow-up for less than 10 years. Additionally, elevated levels of Tregs can help predict inferior OS and PFS in patients with MM (52). Giannopoulos et al. divided patients into two groups based on the median Treg frequency. The OS of patients with lower Treg frequency (< 6.16%) was shorter than in those with high Treg frequency (≥ 16%) (23). Similarly, Muthu Raja KR1 et al. found that patients with ≥ 5% Treg cells had a worse TTP. Univariate Cox regression model analysis showed that only PB Treg cells showed a prognostic role (24). To understand and ultimately utilize the aspects of the immune regulatory mechanism in transplanted MM patients, Franssen LE1 and colleagues retrospectively studied Tregs in 53 MM patients. The relationship between patients with the highest quartile of Treg levels (> 14.6% CD4⁺ T cell subsets) was significantly reduced in PFS and OS. The results of multivariate Cox regression analysis of OS indicate that Treg levels are an independent predictor of OS. High Treg levels exert a negative impact on OS (25). However, a study evaluated the expression in Tregs and Th17 cells of related markers such as FOXP3, CTLA4, and RORγt by performing quantitative real-time PCR. The expression of FOXP3 and CTLA4 in MM patients was found to be 6-fold and 30-fold higher, respectively, than that in the control group. There was no significant difference in the expression of RORγt and other genes related to Treg and Th17 cell subsets. Further univariate analysis showed that none of the CD4+ T cell-associated genes exerted an effect on the prognosis of patients (53). This might be related to the screening procedures and the number of samples included.

Natural killer cells constitute an innate lymphocyte response to cancer and viral infections (54). Natural killer cell-mediated immune function is further deteriorated in advanced MM cases. Compared with MGUS and untreated MM, the number of PB NK cells in advanced disease is substantially reduced (55). NK cells remain functional in patients with MGUS and, during the progression of MM, NK cell function may notably alter and eventually inhibit the development of advanced disease. Additionally, NK cell activity is positively correlated with disease-free survival in patients with MM (26).

NK cell count is a predictive index for MM survival. MM patients, one month after allogeneic hematopoietic stem cell transplantation, with NK cell counts below 100 cells/mL show shorter PFS than patients with NK cell count ranging between 100 and 200 or with counts above 200 cells/mL (2.2 vs. 11.6 months) (27). NK cells are significantly more abundant in PB in MM patients with long-term disease than those in healthy donors or MGUS (49).

However, For the function of NK cell,NK cell-activating receptors such as natural killer group 2D (NKG2D), NKp30, and NKp44 have no evident prognostic prediction value.

Gamma delta (γδ) T cells, which are innate immune cells, play an important role in anti-tumor immune surveillance (56). However, there are no significant differences in PB or BM γδ-T cell counts between MGUS and MM patients compared to those in healthy controls (49). Another study enrolled 101 MM patients and found that higher γδT cell, and CD4 ⁺ central memory (CM) cell counts after AHSCT approximately 100 days were related to a superior 2-year OS (20).

MDSCs are an important cell type described as a heterogeneous subset of immature myeloid cells (57). These cells are an important part of the MM micro environment and modulate MM cell survival and immune escape. For G-MDSCs, Giallongo et al., Brown et al., and Indu et al. evaluated their abundance in the PB or BM of MM patients with newly diagnosed or relapsed cases, and found that the abundance was significantly higher when compared to patients with MM in remission, MGUS, and healthy subjects (57–60). Binsfeld et al. obtained similar results in murine MM models (61). In other studies conducted, M-MDSCs (CD14⁺ monocytic) were regarded as more important than G-MDSCs (CD15⁺ granulocytic) (62). Wang et al. and Anderson et al. reported that the abundance of M-MDSCs in PB or BM in newly diagnosed and relapsed MM patients were significantly increased compared with those in MM patients in remission as well as those in healthy donors (63). More importantly, they revealed that M-MDSC count was significantly related to disease activity and tumor progression (64). Lee et al. also revealed that the increase in peripheral M-MDSCs was associated with failure to achieve a response of VGPR or greater, suggesting poorer efficacy of the therapy (65). Min et al. and LE et al. demonstrated that pre-ASCT M-MDSC burden was correlated with a higher ISS stage and a lower TTP after ASCT (25, 28).

Macrophages contribute to the development of MM-associated neovascularization through both the paracrine secretion of angiogenic factors (angiogenic pathway) and a vasculogenic pathway, and may therefore represent a significant target for conducting anti-neovessel treatments in MM (54). MM cells can influence the polarization of macrophages by increasing the expression of M2-related scavenger receptors. Additionally, macrophages that are modulated by MM cells can inhibit the proliferation of T cells and the production of IFNγ  (66).

BMSCs play a vital role in the development of MM patients. It has been revealed that the proliferative capacity of MM patient-derived MSCs is lower than that in healthy donors, accompanied by a decreased expression of MSC-related receptors (67). Other studies revealed that cross-talk established between BMSCs and MM cells can support the proliferation of myeloma cells by the secretion of IL-6, cell growth factor, and other factors (68, 69). For important fact, BMSC can down regulate immune function, such as BMSC can negatively regulate the CTL cell by PD-1/PDL-1 pathway in our previous research (70).

OCs increase the expression of immune checkpoints such as CD200 (CD200R), Galectin-9 (Tim-3), and PD-L1 (PD-1) during osteoclastogenesis, which can inhibit the activity of T cells by the immune checkpoint (71). Furthermore, OCs can secrete Galectin-9 and APRIL. Galectin-9 induces the apoptosis of T cells and leads to an increase in MM cells, and APRIL can induce the expression of PD-L1 in MM cells and aid their immune escape (72, 73).

It has been reported that OBs are related to the inhibition of MM immunity (74). Myeloma cells inhibit the differentiation and maturity of OBs (75), and the production of cytokines, chemokines, and inflammatory factors in the bone marrow micro environment is involved in the regulation of both cells (76).

Although there lot of evidence indicated that those cell involved in the regulation of immune function in MM patients. The roles in predicting prognosis of MM patients remain unknown owing to insufficient study and a long time follow up clinical data. Further long-term follow-up studies will provide evidence to support the utilization of these markers in MM clinical prognosis prediction.

Cytokines can secrete by immune cell and other type cell such as tumor cell, Endothelial cell and even smooth muscle. It can be divided into inflammatory (such as IL-1 and IL-6) and anti-inflammatory (such as IL-1Rα, IL-4, and IL-10) types based on their effects on disease progression (77). Different cytokines play various roles in patients with MM. Here we focus on some major cytokines secreted by immune cell in MM.

Complex class I-related chain molecule A (MICA) is a ligand for NKG2D. A previous study revealed that the expression of MICA in plasma cells was related to the progression of MGUS to MM, and owing to a high expression of MICA in MGUS plasma, the immune response of NKG2D+ lymphocytes, such as NK cells, γδ-T-cells, and CTL cells (91, 92), could be induced.

Some tumor cell associated cytokines such as cytokines and angiogenic factors (CAFs)which can support the proliferation of tumor cell.FGF-2, HGF, VEGF, and PDGF-β plasma levels at diagnosis are indicative of more profound response since lower angiogenesis to MM cell. Furthermore, MM patients with low levels of FGF-2, VEGF showed superior PFS (81).

MM cell can also secrete cytokines.IL-6 plays a vital role in the proliferation of MM cells (79). It has been reported that high levels of IL-6 are related to disease progression. MM patients with high IL-6 levels (> 7 pg/mL) show inferior survival compared to patients with low levels (<7 pg/mL) (2.7 vs. 53.7 months) (80). Since IL-6/STAT3 signaling can promotes the creation of angiogenesis via enhancement of VEGF, the stimulation by IL-6/STAT3 signaling can also active MM proliferation related pathway like Ras, Akt and MAPK (93).

Other non-cellular components such as complement (94) and adiponectin (95) have been reported to be correlated with the development of MM. However, there are no data supporting its prognostic value.