Prognostic features of multiple myeloma

Best Practice & Research Clinical Haematology Vol. 18, No. 4, pp. 569–583, 2005 doi:10.1016/j.beha.2005.01.012 available online at http://www.sciencedirect.com

6 Prognostic features of multiple myeloma Jesu´s F. San Miguel* MD, PhD Professor

Ramo´n Garcı´a-Sanz

MD, PhD

Consultant Department of Hematology, University Hospital of Salamanca, Paseo de San Vicente, 58-182, Salamanca 37007, Spain

The outcome of myeloma patients is highly heterogeneous, with survival ranging from a few months to more than 10 years. Accordingly, investigation of prognostic factors may contribute to identification of risk categories and to provision of more accurate information about individual disease outcome. For many years prognostic factors have relied on clinical parameters such as age, hemoglobin level and renal function. Subsequently, biological parameters such as the proliferative activity of plasma cells and b2-microglobulin have been added to the prognostic arsenal. More recently, cytogenetic and molecular markers with significant influence on disease outcome have been identified. Here we will review the most relevant prognostic factors reported in the literature in patients treated with both conventional chemotherapy and high-dose therapy followed by autologous stem-cell support, as well as in asymptomatic MM and MGUS patients. Key words: multiple myeloma; prognosis; treatment outcome; treatment failure; outcome assessment.

The outcome of myeloma patients is highly heterogeneous, with survival ranging from a few months to more than 10 years. This is the main reason for extensive investigation on prognostic factors, so more than 50 parameters related to prognosis have been described. However, there is a disassociation between the large body of information generated in this field and its translation into clinical practice. Investigation on prognostic factors is mandatory for three reasons: (1) to provide more individualized information about the disease outcome, something systematically required by the patients and their families; (2) to identify risk groups in order to adapt patient treatment according to the expected outcome and to compare the results between different * Corresponding author. Tel.: C34 23 291384; Fax: C34 23 294624. E-mail address: [email protected] (J.F. San Miguel). 1521-6926/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

570 J. F. San Miguel and R. Garcı´a-Sanz

therapeutic strategies; and (3) to discover associations between biological features of the tumor clone and clinical behavior in order to better understand the pathogenic background of the disease. In this review, we will discuss the most relevant prognostic factors reported in the literature, which have usually been identified upon analyzing symptomatic multiple myeloma (MM) patients.1–5 We will then review other situations where evaluation of prognostic factors is of help, such as asymptomatic MM and patients undergoing highdose therapy. There are three main categories of prognostic factors: (1) factors depending on the host; (2) factors that reflect specific characteristics of the malignant clone; and (3) factors resulting from the interaction between the tumor clone and the host, which mainly reflect the tumor burden and disease complications.

PARAMETERS RELATED TO THE PATIENT The favorable influence of a good performance status and young age are well-established features among host-dependent factors. An age less than 60–70 years is associated with prolonged survival.1–5 Moreover, it has been reported that patients under 40 years of age with normal renal function and low b2M have a median survival of over 8 years.6 In addition, very advanced age is also a poor prognostic factor; thus, we have observed that patients over 80 years of age have a much worse prognosis than patients between 65 and 80 years of age, independently of other prognostic factors.7 Similarly, the presence of a good performance status—close to normal, for instance an Eastern Cooperative Oncology Group (ECOG) scale of 0–2—confers a better prognosis for the patient.5 Other factors within this group, such as sex or race, are of lower value. Thus, the Southwest Oncology Group published a study showing a similar survival in black and in white patients.8 No similar studies comparing other ethnic groups have yet been developed. The immune status of the patient plays an important role in the control of tumor growth. We have observed that the number of CD4 cells is significantly reduced in MM patients, particularly in those in advanced clinical stages, and that the reduction is mainly due to memory and not to naive CD4 cells. Moreover, patients with low CD4 levels (!700!106 cells/L) display a significantly shorter survival, although this is not an independent prognostic factor.9 Similar results have been reproduced by the ECOG.10 In addition, this group observed that high levels of peripheral-blood CD19C cells are positively associated with prolonged survival.11 Within NK cells there are several NKcell subpopulations with antigenic and functional differences, which has generated discrepant results.12,13 In our experience, the overall number of NK cells is significantly increased in patients with MM, but their distribution according to clinical stages differs depending on the type of NK subpopulation. Thus, the number of mature NK cells increases in the early stages of the disease, probably in an attempt to control the tumor growth, while in advanced stages the number of peripheral-blood mature NK cells decreases, although the relative number of immature NK cells increases.13 Another important finding concerning the prognostic influence of immune surveillance is that MM patients might develop T-cell clones which can recognize autologous idiotypic Ig structures as tumor-specific antigens. The occurrence of expanded T-cell cytotoxic clones within the CD8CCD57CCD28K compartment is associated with an improved prognosis.14,15

Prognostic features of multiple myeloma 571

FACTORS RELATED TO THE TUMOR-CELL CLONE This group of prognostic factors reflects intrinsic characteristics of the myelomatous plasma cells, including their morphology, immunocytochemistry, immunophenotyping, cytogenetics, oncogenes and gene expression, multidrug resistance and the proliferative activity. In contrast to other hematological malignancies such as acute leukemias and lymphomas, where the malignant cell pathology is quite significant for classification and prognosis, little attention has been paid to the morphological characteristics of plasma cells. Nevertheless, an immature or plasmablastic morphology has been shown to be associated with a poor outcome and has independent prognostic significance.16–18 In addition, the Southwest Oncology Group showed that low plasma-cell acid phosphatase levels are associated with poor survival (1.7 years for patients with low scores versus 2.8 years for those with high scores).19 The prognostic influence of immunophenotyping has been extensively investigated20–24, but in summary CD20 and sIg expression (identifying immature plasma cells) tend to be associated with poor prognosis.20,21 Moreover, down-regulation of CD56 and CD11a and higher expression of CD44 have been associated with extramedullary spreading of malignant plasma cells22,23, and CD28 expression is probably related with a highly proliferative accelerated phase of the disease.22,25 In any case, the prognostic impact of these antigens has not been confirmed as an independent prognostic factor, so routine immunophenotyping should not be performed for prognostic evaluation purposes. In contrast, these studies can be of help for differential diagnosis between MGUS and MM26,27 as well as for monitoring minimal residual disease following treatment. The CD138 antigen (syndecan-1) is shed from the surface of plasma cells and can be measured in serum. The Nordic Myeloma Study Group has reported that patients with high serum syndecan-1 (R1170 units/mL) display a short survival (20 versus 44 months).28 As in other hematological malignancies, cytogenetics has became one of the most important prognostic findings for MM. Therefore, cytogenetic evaluation is now mandatory in all patients with newly diagnosed MM. The Arkansas Group29,30 has shown that, in MM patients undergoing high-dose chemotherapy followed by autologous stem-cell transplantation, either partial or complete deletions of chromosome 13 are associated with shorter survival. The adverse influence of K13q has also been reported for patients treated with conventional chemotherapy by several other groups, including our own.31–33 It has been suggested that the prognostic impact of monosomy 13 is superior when detected by conventional cytogenetics as compared to detection by fluorescence in situ hybridization (FISH), since in the former, plasma cells should display not only the cytogenetic abnormality but also a proliferative capacity high enough to yield good quality mitosis. Nevertheless, there have been several papers in which the detection of monosomy 13 by FISH is of prognostic relevance for patients treated either with standard or high-dose therapy31,34, and this technique has become the standard for detection of 13q deletions. There are other chromosomal changes that have been reported to be associated with poor survival, including t(4;14)(p16;q32), t(14;16)(q32;q23) and 17p13 deletion (p53).35,36 In contrast, the presence of t(11;14)(q13;q32) appears to have a favorable prognosis.37 In addition, the presence of complex as well as hypodiploid karyotypes is also associated with treatment failure. Other potential adverse cytogenetic features are chromosome 1p/1q abnormalities and chromosome 22 deletions.34,38 By contrast,

572 J. F. San Miguel and R. Garcı´a-Sanz

trisomies of chromosomes 9, 11 and 17 tend to have a favorable prognostic influence.31 In line with this latter observation, the presence of a DNA cell content with a DNA index O1 is associated with a significantly better prognosis than those with a DNA index %1, although without a significant independent prognostic value.39 A recent study carried out by our group, based on comparative genomic hybridization, has shown that chromosome losses (not only K13q) and gains on 11q are associated with poor prognosis.40 Alteration in oncogene and tumor suppressor gene expression has become an area of increasing interest for clarifying the relationship of such alterations with the pathogenesis of the disease and, therefore, with clinical behavior and outcome. There is general agreement that tumor clone development and progression are consequences of a multi-step process that accumulates sequential genetic changes. Some of these genetic abnormalities have been found to have an impact on the aggressiveness and prognosis of the disease. P53 mutations41 or deletions42 are associated with progressive disease and relapse, and therefore could reflect disease aggressiveness and treatment refractoriness. This adverse influence has also been observed in patients treated with high-dose therapy.43 The retinoblastoma (Rb) gene is located on chromosome 13, and accordingly, the adverse prognostic impact of Rb deletions parallel that of monosomy 13.31,33,42 Patients who display K-ras mutations have significantly shorter survival as compared to those who do not (2 versus 3.7 years).44 In addition, our group observed that methylation of the p16 suppressor gene is associated with high proliferative activity of plasma cells and poor prognosis, but it is not an independent prognostic factor because of its relationship with the number of plasma cells in Sphase.45 This association between p16 methylation and poor prognosis has also been found by others.46–48 The availability of technical platforms that allows multi-evaluation of gene expression (high-density oligonucleotide DNA microarray) has permitted the identification of previously unrecognized alterations in oncogene and tumor suppressor gene expression.49–52 Accordingly, the Arkansas group has proposed a classification for MM patients into four different molecular subtypes (MM1, MM2, MM3 and MM4) based on different gene expression profiles, which are correlated with specific biological characteristics, chromosome alterations, and clinical outcome.53 MM1 patients have the best prognosis while MM4 have the worst. The counterpart of the most undifferentiated MM (MM4) would be tonsil B cells, for MM3 tonsil plasma cells, MM2 would correspond to bone-marrow plasma cells, while for MM1 the normal counterpart has not yet been identified. Nevertheless, the real clinical value of this classification remains to be established. Moreover, these four different groups of diseases initially proposed will probably be redefined to include up to seven different types. An alternative classification combines the information derived from chromosomal translocations and cyclin D expression, the so-called TC classification (T for translocations and C for cyclin D) that divides MM patients into five groups: TC1 tumors, expressing high levels of either cyclin D1 or cyclin D3 as a result of an Ig translocation [t(11;14 or t(6;14)]; TC2 tumors, which ectopically express low to moderate levels of cyclin D1 in the absence of a t(11;14) translocation; TC3 tumors that are a group of tumors identified by exclusion (they do not fall into one of the other groups), with most expressing cyclin D2, but a few also expressing low levels of cyclin D1 or D3; TC4 tumors, expressing high levels of cyclin D2, and also MMSET/FGFR3 as a result of a t(4;14); and TC5 tumors express the highest levels of cyclin D2, and also high levels of either c-maf or mafB. The relative percentages of each of these different

Prognostic features of multiple myeloma 573

tumors are: 18, 37, 22, 16 and 7%, respectively. However, the translation of this classification into the prognosis has not been totally established yet.54 There are some parameters that reflect the global effect of some of these genetic alterations all together. We are referring to the assessment of the proliferative activity of the malignant plasma cells by techniques such as labeling index (LI) with bromodeoxyuridine55 or by flow cytometry with propidium iodide (PI).56 This has been considered as one of the most important prognostic markers for MM. In our experience, the number of plasma cells in S-phase, together with cytogenetics, b2M, PS and age, represents the best combination of disease characteristics for survival prediction.24 In this line, the Australian group57 has shown that the proliferative activity is almost entirely attributable to an increase in the LI of the primitive plasma cells. Interestingly, we have also observed that a high proliferative plasma cell activity is a deleterious factor for survival independent of age, so it is also the most important prognostic factor for survival in patients over 65 years of age.7 An interesting factor related to prognosis is the ability of tumor cells to overcome the toxic effect of chemotherapeutic drugs. Several approaches have been used to assess it, such as the expression of p-glycoprotein (MDR-1) which is usually observed in patients who have been exposed to anthracyclines and vinca alkaloids58,59, although its value as a factor has yielded discrepant results.60,61 A more interesting approach is the assessment of LRP (lung resistant protein), since it has been reported that this protein is expressed in 50% of MM patients, and it identifies a subgroup of patients with a low probability of responding to conventional chemotherapy but who may respond to high-dose chemotherapy and autologous stem-cell transplantation (ASCT).61 It is also worth mentioning that response to front-line therapy is an important prognostic factor.62 In addition, the quality of the response at the end of the therapy is also one of the best predictors for survival in MM patients.63 However, we should be aware that patients with a rapid response may have a short duration of response and survival. This was suggested 25 years ago64 and subsequently confirmed with kinetic studies65, but this statement is under review, since some authors have recently found that a response better than 30–40% M-component decrease after the first cycle of melphalan would be associated with a very good prognosis.66 Moreover, there are now more players in the game, since the South West Oncology Group has recently published that the time to first progression is a more important factor than any type of response.67

PROGNOSTIC FACTORS ASSOCIATED WITH TUMOR BURDEN AND DISEASE COMPLICATIONS The information in this group of prognostic factors is the most abundant and can be distributed into three subgroups: (1) factors related to the malignant clone expansion/tumor burden; (2) factors related to disease complications—anemia, renal insufficiency, skeletal lesions, etc; and (3) biochemical markers, including some cytokines, that reflect disease activity. The first attempt to measure the tumor burden was made through the Durie– Salmon classification, which was obtained by using mathematical models that mainly tried to establish a relationship between the tumor mass and the M-component size. Other features that may reflect the tumor burden are the proportion of plasma cells in

574 J. F. San Miguel and R. Garcı´a-Sanz

bone marrow, the bone-marrow pattern of infiltration, and the presence of circulating plasma cells. Consequently, a high number of plasma cells in bone marrow, as well as a diffuse pattern of infiltration, are generally associated with a poor prognosis. However, these are not consistent prognostic factors, probably due to the heterogeneous distribution of plasma cells in bone marrow (the areas with bone tenderness or lytic lesions are usually more heavily infiltrated). The detection of circulating plasma cells, identified either by morphology or by immunophenotyping, is associated with advanced disease, and it has been reported that the presence of high levels of circulating plasma cells (O4%) is an independent adverse prognostic factor.7,68 Many disease complications (anemia, thrombocytopenia, renal insufficiency) have a relevant influence on disease outcome.69 Skeletal lesions also have an adverse impact on survival. They have usually been evaluated by X-ray and, more recently, by bone resorption markers such as pyridinoline (PYD) and deoxypyridinolin (DPD). The urinary levels of PYD and DPD are augmented in patients with advanced clinical stages and progressive disease, and they are correlated with C-reactive protein (CRP), creatinine and albumin levels, but the relationship with survival is not very close.69,70 Several biochemical markers, including some cytokines linked to disease activity, afford relevant prognostic information. The most important of them is the uncorrected b2M levels that increase as a result of both tumor burden growth and renal function deterioration. Actually, it is a very sensitive indicator of glomerular filtration. Thus, this parameter is related to any variable that affects renal function such as the myeloma impact on the kidney, age, infections or other kidney problems. This is the main reason why b2M abrogates the independent prognostic value of the serum creatinine, urea and age in multivariate analyses. Several threshold values for b2M (3–6 mg/dL) have been used to discriminate prognostic subgroups24,71–74, but in our experience b2M levels can also be used as a continuous variable since the higher the b2M value, the shorter the survival. This finding makes mandatory the assessment of serum b2M levels at diagnosis in MM. However, it is not helpful for monitoring the course of the disease, since there are patients relapsing with or without previous increases in b2M levels.74 Other markers of disease activity—such as thymidine kinase, neopterin and lactate dehydrogenase—do not usually remain as independent prognostic factors in multivariate analyses.69 Interleukin-6 (IL6) is a major plasma cell growth factor, and elevated serum levels have been associated with short survival.75,76 Similarly, high levels of its soluble receptor (sIL-6R) correlate with poor prognosis.7,77 Nevertheless, discrepant results have also been reported78, so these two markers are not extensively used in clinical practice to asses the prognosis in MM patients. Moreover, IL-6 influences the hepatic synthesis of several acute-phase reactant proteins such as CRP, a1-antitrypsin (a1AT) and orosomucoide that could be used in its place. Actually, serum CRP levels represent a surrogate marker for IL-6 concentration79, and the same could be applied to a1AT.80 In addition, the assessment of these proteins is very simple and cheap, therefore either a1AT or particularly CRP can replace IL-6 as prognostic factors.4 The use of CRP together with b2M constitutes a very useful combination for predicting survival in MM patients, allowing stratification of MM patients into three groups according to CRP and b2M serum levels: (1) low-risk group, CRP and b2M !6 mg/L (50% of patients); (2) intermediate risk group, CRP or b2M R6 mg/L (35% of patients); (3) high-risk group, CRP and b2M R6 mg/L (15% of patients). Survival was 54, 27, and 6 months, respectively (P!0.0001).79 The South West Oncology Group has proposed a similar scheme by substituting CRP with albumin levels. Accordingly, stage I, II and III patients have albumin R3 g/dL with different possibilities in the level of b2M

Prognostic features of multiple myeloma 575

(!2.5, 2.5–5.5 and O5.5 mg/L, respectively), while stage IV corresponds to cases with albumin !3 g/dL and b2M O5.5 mg/L. In the previous pages, we have discussed a large number of prognostic factors, but perhaps only a few of them have real independent value. A summary of the most important would include: (1) two host factors that reflect the ability of the patient to tolerate chemotherapy (age and performance status); (2) two intrinsic characteristics of the malignant clone (cytogenetics and proliferative activity LI) together with two biochemical markers that reflect tumor burden/disease complications (b2M and CRP).

PREDICTING TRANSFORMATION IN ASYMPTOMATIC MYELOMA AND MONOCLONAL GAMMOPATHY OF UNDETERMINED SIGNIFICANCE (MGUS) The number of patients diagnosed with asymptomatic (smouldering) MM has increased in recent decades. Within asymptomatic MM patients it is of great interest to identify those cases at high risk of progression, since they might benefit from early treatment. The study conducted by Facon et al80 included 91 stage I asymptomatic MM patients, of whom 41 experienced disease progression at a median of 48 months. In multivariate analysis, the only significant factors influencing progression were: Hb !12 g/dL, bonemarrow plasmacytosis O25% and MC O30 g/L. Patients with two or three risk factors progressed within 6 months while those without risk factors remained free of progression for O50 months. A second study, conducted at MD Anderson Cancer Center81, included 101 asymptomatic patients without lytic lesions. In this series, the presence of two out of the following three factors identified a subgroup of patients at high risk of early progression: M-component O30 g/L, BJ protein O50 g/L and IgA isotype. In addition, in patients with only one adverse factor, magnetic resonance imaging (MRI) proved to be a useful tool for the identification of cases at risk of progression. The present data illustrate how these types of study may help physicians in making clinical decisions for a difficult subset of MM patients in which treatment is controversial. Upon considering MGUS patients, the most extensive study was published in 2002 by Kyle et al82 It included 1384 patients followed from 1960 through 1994 covering 11 009 person-years of follow-up and 115 progressions to multiple myeloma, IgM lymphoma, primary amyloidosis, macroglobulinemia, chronic lymphocytic leukemia, or plasmacytoma. The risk of progression to MM or a related cancer was related only to the size of the M-component at diagnosis. Thus, 10 years after the diagnosis of MGUS, the risk of this progression was 6% for an initial monoclonal protein value %0.5 g/dL, 7% for a value of 1 g/dL, 11% for 1.5 g/dL, 20% for 2 g/dL, 24% for 2.5 g/dL, and 34% for O3.0 g/dL (P!0.001). Further studies based on cytogenetic/molecular parameters are needed in order to improve risk progression assessment in this cohort of patients.

AUTOLOGOUS AND ALLOGENEIC STEM-CELL TRANSPLANTATION The extensive use of ASCTas a therapeutic option for a large number of MM patients has stimulated the evaluation of prognostic factors in this setting. Although ASCT is associated with prolongation in both survival and quality of life, because of its high cost it

576 J. F. San Miguel and R. Garcı´a-Sanz

is important to have models to predict which patients will really benefit from transplants, thereby avoiding the financial and emotional burden imposed on those unlikely to respond. In this area, it would also be desirable to have predictive models to individualize the choice between allogeneic and autologous transplants in young patients. The first question on prognostic factors for patients undergoing high-dose chemotherapy is whether the influence of prognostic parameters changes from that observed under conventional chemotherapy. Current data suggest that they are similar. According to the first observations made in Arkansas29,83, low b2M levels, less prior therapy and Ig isotype different from IgA were of favorable prognostic value for overall survival (OS) and disease-free survival (DFS). Later, the same group added information on cytogenetics, showing that chromosome 11q and 13 abnormalities were associated with poor outcome in patients receiving tandem autologous transplants: event-free survival (EFS) (21 versus 50 months PZ0.0001) and OS (34 versus C62 months, PZ0.001).32 Moreover, the presence of any type of translocation (unfavorable-complex karyotype) is a dominant adverse prognostic variable.32 As previously mentioned, the adverse impact of p53 also applies to patients treated with high-dose chemotherapy. Similar results were found at the European Bone Marrow Transplantation (EBMT) study84, since stage I, pre-transplant complete remission (CR), one line of therapy, age !45 years, and low b2M levels were all favorable factors. Controversial results exist concerning the influence of LI, since some groups85 have suggested that its adverse effect can be overcome with high-dose therapy, while others86 have observed that a high LI as well as the presence of circulating monoclonal plasma cells in the blood stem cell harvests were associated with shortened survival after transplantation. Interestingly, in both studies, b2M retained its adverse prognostic influence. All these considerations suggest that most prognostic factors for conventional chemotherapy retain the prognostic influence under high-dose therapy. Nevertheless, this is still an open area, since the favorable outcome of patients with t(11;14) is clearer in patients receiving ASCT than conventional chemotherapy (predicted 88% OS at 80 months).54 The use of high-dose chemotherapy has modified the concept of response in MM, since complete responses are now a possibility that was infrequently observed under conventional chemotherapy. Moreover, several studies have shown that achievement of CR—defined by electrophoresis and immunofixation—is the most important prognostic factor in MM.63,87–89 More sensitive techniques such as multiparametric immunophenotyping and molecular biology (reverse transcriptase polymerase chain reaction, RT-PCR) may additionally contribute to a better redefinition of risk group categories. Thus, using molecular techniques, we have shown90 that MM patients achieving a CR or near CR in whom the presence of less than one tumor cell per 104 total bone-marrow cells display a very good outcome. Very similar data have been reported by other groups with very similar cut-off points.91,92 Our own group obtained parallel results by using flow cytometry in a series of 86 patients that had achieved CR or near CR after high-dose therapy.93 One interesting additional issue is to examine variables associated with long-term (R5 years) EFS after transplant. The Arkansas group94 reported that the most favorable features are pre-transplant CRP !4 mg/L, b2M !2.5 mg/dL, prior therapy %12 months and absence of adverse cytogenetic abnormalities. Regarding response to transplant, although attaining a CR was a favorable variable in univariate analysis, it was not an independent feature in multivariate analysis.

Prognostic features of multiple myeloma 577

In the allogeneic BMT setting the experience of Seattle95 and EBMT96 shows that b2M is again the most important prognostic factor, together with clinical stage, age, and lines of treatment before transplantation. It should be noted that neither the influence of the LI nor cytogenetics was evaluated in any of these studies. More recently information on prognostic factors in patients receiving allogeneic stem-cell transplantation with a reduced-intensity regimen have became available. Here again deletion of the chromosome 13q1497 had an independent influence on disease outcome.98

AN INTERNATIONAL PROGNOSTIC INDEX FOR MYELOMA Since the introduction of ASCT as well as new drugs such as thalidomide, bortezomib and ImiDs, among others, clinicians have a wide array of therapeutic tools for MM patients. It would be desirable, as in other disorders such as acute lymphoblastic leukemia and non-Hodgkin’s lymphoma, to have an international classification system for MM patients based on risk group categories, defined according to new independent prognostic factors. This system would assist us to individualize treatment according to patients’ characteristics and to help in clinical decision-making. In addition, prognostic factors could represent a valuable tool for the evaluation of results of new treatment strategies. In this context, experimental therapies should be assayed on homogenous cohorts of patients identified according to prognostic factors. Moreover, upon evaluating randomized trials, for the balance of the two arms it would be important to take into account not only the individual prognostic factors but also the possible additive effect of two or three prognostic factors within a particular therapeutic arm. According to these statements, a new international staging system (ISS) has recently been proposed. It derives from a multicenter study collecting a total of 11 171 patients compiled from American, Asian and European cooperative groups and large individual institutions.99 The ISS is based on the levels of b2M and albumin (Table 1), and it allows delineation of three risk groups regardless of age, geographic region or standard or transplant therapy. Further efforts are urgently needed in order to improve this staging system with the inclusion of other parameters such as cytogenetics and molecular markers. Although this classification will not be widely applicable, since these later parameters are only available at selected centers, it may be of great value in the evaluation of new treatment strategies at large cooperative groups and referral institutions.

SUMMARY The survival duration of patients with multiple myeloma ranges from months to more that 10 years. In the main, this heterogeneity relates to specific characteristics of Table 1. International staging system (ISS) for multiple myeloma. Stage 1 2 3

Parameters

Median survival (months)

b2M !3.5 mg/L and albumin O3.5 g/dL b2M !3.5 mg/L and albumin !3.5 g/dL, or b2M 3.5–5.5 mg/L b2M O5.5 mg/L

62 44 29

578 J. F. San Miguel and R. Garcı´a-Sanz

the tumor itself and the host. Many of these characteristics are interrelated, and in fact only a few of them remain as independent prognostic factors in multivariate analysis. These latter factors can be systematized into three categories: (1) host factors that reflect the ability of the patient to tolerate chemotherapy (age, performance status, renal function, etc.); (2) intrinsic characteristics of the malignant clone (cytogenetics, molecular profile, proliferative activity of tumor cells, etc.); and (3) biochemical markers that reflect tumor burden and disease complications (b2M, albumin, etc.). Gene expression profiling studies are emerging as most relevant in this field, and hopefully will contribute to the identification of those markers that play a critical role in MM pathophysiology and outcome. The identification of those characteristics associated with either a good or a poor prognosis will help to stratify patients according to risk groups and to adapt treatment to the expected outcome (i.e. treatment strategies associated with a high co-morbidity would be restricted to high-risk patients). This is particularly important upon evaluating new therapeutic strategies. The international staging system which is based on two widely available parameters (b2M and albumin) is a first step in a joint effort that has included the largest cooperative groups and referral institutions, and should continue to implement the accuracy of these classifications through the inclusion of novel parameters such as cytogenetic and molecular markers. Practice points † † † † † † † †

Prognostic factors include: creatinine b2M CRP hemoglobin lactate dehydrogenase calcium cytogenetics FISH (Rb, IgH translocations)

Research agenda † † † † †

FISH (chromosomal abnormalities not previously investigated for prognosis) labeling index/cell cycle analysis gene profiling by arrays gene expression by RT-PCR MRD investigation both by multiparametric flow cytometry and PCR

REFERENCES 1. Kyle RA. Why better prognostic factors for multiple myeloma are needed? Blood 1994; 83: 1713–1716. 2. Kyle RA. Prognostic factors in multiple myeloma. Stem Cells 1995; 13: 56–63.

Prognostic features of multiple myeloma 579 3. San Miguel JF. Overview of prognostic factors in multiple myeloma. Cancer Res Ther Control 1998; 6: 97–99. 4. Turesson I. Prognostic evaluation in multiple myeloma: an analysis of the impact of new prognostic factors. Br J Haematol 1999; 106: 1005–1012. 5. San Miguel JF & Garcı´a-Sanz R. Multiple myeloma: differential diagnosis and prognosis. In Gahrton G, Durie DGM & Samson Dm (eds.) Multiple Myeloma and Related Disorders. Oxford University Press, New York, 2004, pp. 179–199. 6. Blade´ J. Presenting features and prognosis in 72 patients with multiple myeloma who were younger than 40 years. Br J Haematol 1996; 93: 345–351. 7. Garcı´a-Sanz R, Gonza´lez-Fraile MI, Mateo G et al. Proliferative activity of plasma cells is the most relevant prognostic factor in elderly multiple myeloma patients. Int J Cancer 2004; 112: 884–889. 8. Modiano MR, Villar-Werstler P, Crowley J & Salmon SE. Evaluation of race as a prognostic factor in multiple myeloma. An ancillary of Southwest Oncology Group Study 8229. J Clin Oncol 1996; 14: 974–977. 9. San Miguel JF, Gonza´lez M, Gasco´n A et al. Lymphoid subsets and prognostic factors in multiple myeloma. Cooperative Group for the Study of Monoclonal Gammopathies. Br J Haematol 1992; 80: 305–309. 10. Kay NE, Leong TL, Bone N et al. Blood levels of immune cells predict survival in myeloma patients: results of an Eastern Cooperative Oncology Group phase 3 trial for newly diagnosed multiple myeloma patients. Blood 2001; 98: 23–28. 11. Kay NE, Leong T, Kyle RA et al. Circulating blood B cells in multiple myeloma: analysis and relationship to circulating clonal cells and clinical parameters in a cohort of patients entered on the Eastern Cooperative Oncology Group Phase III E9486 Clinical Trial. Blood 1997; 90: 340–345. ¨ sterborg A, Nilsson B & Bjo¨rkholm M. Natural killer cell activity in monoclonal gammopathies: relation 12. O to disease activity. Eur J Immunol 1990; 45: 153–157. 13. Garcia-Sanz R, Gonzalez M, Orfao A et al. Analysis of natural killer-associated antigens in peripheral blood and bone marrow of multiple myeloma patients and prognostic implications. Br J Haematol 1996; 93: 81–88. 14. Brown RD, Yuen E, Nelson M et al. The prognostic significance of T cell receptor beta gene rearrangements and idiotype-reactive T cells in multiple myeloma. Leukemia 1997; 11: 1312–1317. 15. Sze DM, Giesajtis G, Brown RD et al. Clonal cytotoxic T cells are expanded in myeloma and reside in the CD8(C)CD57(C)CD28(K) compartment. Blood 2001; 98: 2817–2827. 16. Greipp PR, Leong T, Bennett JM et al. Plasmablastic morphology—an independent prognostic factor with clinical and laboratory correlates: Eastern Cooperative Oncology Group (ECOG) myeloma trial E9486 report by the ECOG Myeloma Laboratory Group. Blood 1998; 91: 2501–2507. 17. Bartl R, Frisch B, Burkhardt R et al. Bone marrow histology in myeloma: its importance in diagnosis, prognosis, classification and staging. Br J Haematol 1982; 51: 361–375. 18. Greipp PR. Advances in the diagnosis and management of myeloma. Semin Hematol 1992; 92(supplement 2): 24–45. 19. Saeed SM, Stock-Novack D, Pohlod R et al. Prognostic correlation of plasma cell acid phosphatase and beta-glucuronidase in multiple myeloma: a Southwest Oncology Group study. Blood 1991; 78: 3281–3287. 20. San Miguel JF, Gonzalez M, Gascon A et al. Immunophenotype heterogeneity of multiple myeloma: influence on the biology and clinical course of the disease. Br J Haematol 1991; 77: 185–190. 21. Omede P, Boccadoro M, Fusaro A et al. Multiple myeloma: ‘early’ plasma cell phenotype identifies patients with aggresive biological and clinical characteristics. Br J Haematol 1993; 85: 504–513. 22. Pellat-Deceunynck C, Bataille R, Robillard N et al. Expression of CD28 and CD40 in human myeloma cells: a comparative study with normal plasma cells. Blood 1994; 84: 2597–2603. 23. Pellat-Deceunynck C, Barille S, Puthier D et al. Adhesion molecules on human myeloma cells: significant changes in expression related to malignancy, tumour spreading and immortalization. Cancer Res 1995; 55: 3647–3653. *24. San Miguel JF, Garcia-Sanz R, Gonzalez M et al. A new staging system for multiple myeloma based on the number of S-phase plasma cells. Blood 1995; 85: 448–455. 25. Robillard N, Jego G, Pellat-Deceunynck C et al. CD28, a marker associated with tumoral expansion in multiple myeloma. Clin Cancer Res 1998; 4: 1521–1526. 26. Harada H, Kawano MM, Huang N et al. Phenotypic difference of normal plasma cells from mature myeloma cells. Blood 1993; 81: 2658–2663.

580 J. F. San Miguel and R. Garcı´a-Sanz 27. Ocqueteau M, Orfao A, Garcia-Sanz R et al. Expression of the CD117 antigen (c-kit) on normal and myelomatous plasma cells. Br J Haematol 1996; 95: 489–493. 28. Seidel C, Sundan A, Hjorth M et al. Serum syndecan-1: a new independent prognostic marker in multiple myeloma. Blood 2000; 95: 388–392. *29. Tricot G, Barlogie B, Jagannath S et al. Poor prognosis in multiple myeloma is associated only with partial or complete deletions of chromosome 13 or abnormalities involving 11q and not with other karyotype abnormalities. Blood 1995; 86: 4250–4256. 30. Tricot G, Sawyer JR, Jagannath S et al. Unique role of cytogenetics in the prognosis of patients with myeloma receiving high-dose therapy and autotransplants. J Clin Oncol 1997; 15: 2659–2666. *31. Perez-Simon JA, Garcia-Sanz R, Tabernero MD et al. Prognostic value of numerical chromosome aberrations in multiple myeloma: A FISH analysis of 15 different chromosomes. Blood 1998; 91: 3366–3371. 32. Zojer N, Konigsberg R, Ackermann J et al. Deletion of 13q14 remains an independent adverse prognostic variable in multiple myeloma despite its frequent detection by interphase fluorescence in situ hybridization. Blood 2000; 95: 1925–1930. 33. Fonseca R, Harrington D, Oken MM et al. Biological and prognostic significance of interphase fluorescence in situ hybridization detection of chromosome 13 abnormalities (delta13) in multiple myeloma: an eastern cooperative oncology group study. Cancer Res 2002; 62: 715–720. 34. Worel N, Greinix H, Ackermann J et al. Deletion of chromosome 13q14 detected by fluorescence in situ hybridization has prognostic impact on survival after high-dose therapy in patients with multiple myeloma. Ann Hematol 2001; 80: 345–348. *35. Fonseca R, Blood E, Rue M et al. Clinical and biologic implications of recurrent genomic aberrations in myeloma. Blood 2003; 101: 4569–4575. 36. Moreau P, Facon T, Leleu X et al. Recurrent 14q32 translocations determine the prognosis of multiple myeloma, especially in patients receiving intensive chemotherapy. Blood 2002; 100: 1579–1583. 37. Fonseca R, Blood EA, Oken MM et al. Myeloma and the t(11;14)(q13;q32); evidence for a biologically defined unique subset of patients. Blood 2002; 99: 3735–3741. 38. Segeren C, Beverloo B, Poddighe J et al. Abnormal chromosomes 1p/q and 13/13q are adverse prognostic factors for the outcome of upfront high-dose therapy in patients with multiple myeloma. Blood 2002; 98(supplement): 159a. 39. Garcia-Sanz R, Orfao A, Gonzalez M et al. Prognostic implications of DNA aneuploidy in 156 untreated multiple myeloma patients. Br J Haematol 1995; 90: 106–112. 40. Gutierrez NC, Garcia JL, Herna´ndez JM et al. Prognostic and biological significance of chromosomal imbalances assessed by comparative genomic hybridization in multiple myeloma. Blood 2004. First Edition Paper, prepublished online July 6, DOI 10.1182/blood-2004-04-1319. 41. Corradini P, Inghirami G, Astolfi M et al. Inactivation of tumor suppressor genes, p53 and Rb1, in plasma cell dyscrasias. Leukemia 1994; 8: 758–767. 42. Drach J, Ackermann J, Fritz E et al. Presence of a p53 gene deletion in patients with multiple myeloma predicts for short survival after conventional-dose chemotherapy. Blood 1998; 92: 802–809. 43. Chang H, Qi C, Yi Q-L et al. P53 gene deletion detected by fluorescence in situ hybridization is an adverse prognostic factor for patients with multiple myeloma following autologous stem cell transplantation. Blood 2004. First Edition Paper, prepublished online August 31, DOI 10.1182/blood-2004-04-1363. 44. Liu P, Leong T, Quam L et al. Activating mutations of N- and K-ras in multiple myeloma show different clinical associations: analysis of the Eastern Cooperative Oncology Group Phase III Trial. Blood 1996; 88: 2699–2706. 45. Mateos MV, Garcia-Sanz R, Lopez-Perez R et al. Methylation is an inactivating mechanism of the p16 gene in multiple myeloma associated with high plasma cell proliferation and short survival. Br J Haematol 2002; 118: 1034–1040. 46. Uchida T, Kinoshita T, Ohno T et al. Hypermethylation of p16INK4A gene promoter during the progression of plasma cell dyscrasia. Leukemia 2001; 15: 157–165. 47. Guillerm G, Depil S, Wolowiec D & Quesnel B. Different prognostic values of p15(INK4b) and p16(INK4a) gene methylations in multiple myeloma. Haematologica 2003; 88: 476–478. 48. Galm O, Wilop S, Reichelt J et al. DNA methylation changes in multiple myeloma. Leukemia 2004; 18: 1687–1692.

Prognostic features of multiple myeloma 581 49. De Vos J, Couderc G, Tarte K et al. Identifying intercellular signaling genes expressed in malignant plasma cells by using complementary DNAarrays. Blood 2001; 98: 771–780. 50. Chauhan D, Li G, Auclair D et al. Identification of genes regulated by 2-methoxyestradiol (2ME2) in multiple myeloma cells using oligonucleotide arrays. Blood 2003; 101: 3606–3614. 51. Magrangeas F, Nasser V, Avet-Loiseau H et al. Gene expression profiling of multiple myeloma reveals molecular portraits in relation to the pathogenesis of the disease. Blood 2003; 101: 4998–5006. *52. Zhan F, Hardin J, Kordsmeier B et al. Global gene expression profiling of multiple myeloma, monoclonal gammopathy of undetermined significance, and normal bone marrow plasma cells. Blood 2002; 99: 1745–1757. *53. Shaughnessy J. Scientific and clinical applications of DNA microarrays—multiple myeloma as a disease model. PHD Mayo Clin Proc 2003; 78: 1098–1109. 54. Hideshima T, Bergsagel PL, Kuehl WM & Anderson KC. Advances in biology of multiple myeloma: clinical applications. Blood 2004; 104: 607–618. *55. Greipp PR, Lust JA, O’Fallon WM et al. Plasma cell labeling index and beta 2-microglobulin predict survival independent of thymidine kinase and C-reactive protein in multiple myeloma. Blood 1993; 81: 3382–3387. 56. Orfao A, Garcia-Sanz R, Lopez-Berges MC et al. A new method for the analysis of plasma cell DNA content in multiple myeloma samples using a CD38/propidium iodide double staining technique. Cytometry 1994; 17: 332–339. 57. Joshua D, Petersen A, Brown R et al. The labelling index of primitive plasma cells determines the clinical behaviour of patients with myelomatosis. Br J Haematol 1996; 94: 76–81. 58. Grogan TM, Spier CM, Salmon SE et al. P-glycoprotein expression in human plasma cell myeloma: correlation with prior chemotherapy. Blood 1993; 81: 490–495. 59. Sonneveld P, Marie JP, Huisman C et al. Reversal of multidrug resistance by SDZ PSC 833, combined with VAD (vincristine, doxorubicin, dexamethasone) in refractory multiple myeloma. A phase I study. Leukemia 1996; 10: 1741–1750. 60. Gieseler F & Nussler V. Cellular resistance mechanisms with impact on the therapy of multiple myeloma. Leukemia 1997; 11(supplement 5): S1–S4. 61. Raaijmakers HG, Izquierdo MA, Lokhorst HM et al. Lung-resistance-related protein expression is a negative predictive factor for response to conventional low but not to intensified dose alkylating chemotherapy in multiple myeloma. Blood 1998; 91: 1029–1036. 62. Blade J, Lopez-Guillermo A, Bosch F et al. Impact of response to treatment on survival in multiple myeloma: results in a series of 243 patients. Br J Haematol 1994; 88: 117–121. *63. Lahuerta JJ, Martinez-Lopez J, Serna JD et al. Remission status defined by immunofixation vs electrophoresis after autologous transplantation has a major impact on the outcome of multiple myeloma patients. Br J Haematol 2000; 109: 438–446. 64. Hansen OP, Jessen B & Videbaek A. Prognosis of myelomatosis on treatment with prednisone and cytostatics. Scand J Haematol 1974; 10: 282–290. 65. Boccadoro M & Pileri A. Cell kinetics of multiple myeloma. Hematol Pathol 1987; 1: 137–142. *66. Schaar CG, Kluin-Nelemans JC, le Cessie S et al. Early response to therapy and survival in multiple myeloma. Br J Haematol 2004; 125: 162–166. 67. Durie BG, Jacobson J, Barlogie B & Crowley J. Magnitude of response with myeloma frontline therapy does not predict outcome: importance of time to progression in southwest oncology group chemotherapy trials. J Clin Oncol 2004; 22(10): 1857–1863. 68. Witzig TE, Gertz M, Lust JA et al. Peripheral blood monoclonal plasma cells as a predictor of survival in patients with multiple myeloma. Blood 1996; 88: 1780–1787. 69. San Miguel JF, Sanchez I & Gonzalez M. Prognostic factors and classification in multiple myeloma. Br J Cancer 1989; 59: 113–118. 70. Pecherstorfer M, Seibel MJ, Woitge HW et al. Bone resorption in multiple myeloma and in monoclonal gammopathy of undetermined significance: quantification by urinary pyridinium cross-links of collagen. Blood 1997; 90: 3743–3750. 71. Cuzick J, De Stavola BL & Cooper EH. Long-term prognostic value of serum B-2-microglobulin in myelomatosis. Br J Haematol 1990; 75: 506–510. 72. Garewal H, Durie BGM, Kyle RA et al. Serum B2-microglobulin in the initial staging and subsequent monitoring of monoclonal plasma cell disorders. J Clin Oncol 1984; 2: 51–57.

582 J. F. San Miguel and R. Garcı´a-Sanz 73. Bataille R, Greiner J & Sany J. Beta-2-microglobulin in myeloma: optimal use for staging, prognosis and treatment: A prospective study of 160 patients. Blood 1984; 63: 468–476. 74. Boccadoro M, Omede P & Frieri R. Multiple Myeloma: beta-2-microglobulin is not a useful follow-up parameter. Acta Haematol 1989; 82: 122–125. 75. Bataille R, Jourdan M, Zhang XG & Klein B. Serum levels of interleukin 6, a potent myeloma cell growth factor, as a reflect of disease severity in plasma cell dyscrasias. J Clin Invest 1989; 84: 2008–2011. 76. Papadaki H, Kyriakou D, Foudoulakis A et al. Serum levels of soluble IL-6 receptor in multiple myeloma as indicator of disease activity. Acta Haematol 1997; 97: 191–195. 77. Pulkki K, Pelliniemi TT, Rajamaki A et al. Soluble interleukin-6 receptor as a prognostic factor in multiple myeloma. Br J Haematol 1996; 92: 370–374. 78. Ohtani K, Ninomiya H, Hasegawa Y et al. Clinical significance of elevated soluble interleukin-6 receptor levels in the sera of patients with plasma cell dyscrasias. Br J Haematol 1995; 91: 116–120. 79. Bataille R, Boccadoro M, Klein B et al. C-reactive protein and beta-2 microglobulin produce a simple and powerful myeloma staging system. Blood 1992; 80: 733–737. 80. Facon T, Menard JF, Michaux JL et al. Prognostic factors in low tumour mass asymptomatic multiple myeloma: a report on 91 patients. The Groupe d’Etudes et de Recherche sur le Myelome (GERM). Am J Hematol 1995; 48: 71–75. 81. Weber DM, Dimopoulos MA, Moulopoulos LA et al. Prognostic features of asymptomatic multiple myeloma. Br J Haematol 1997; 97: 810–814. 82. Kyle RA, Therneau TM, Rajkumar SV et al. A long-term study of prognosis in monoclonal gammopathy of undetermined significance. N Engl J Med 2002; 346: 564–569. 83. Barlogie B, Jagannath S, Vesole D & Tricot G. Autologous and allogeneic transplants for multiple myeloma. Semin Hematol 1995; 32: 31–44. 84. Bjorkstrand B, Ljungman P, Bird JM et al. Autologous stem cell transplantation in multiple myeloma: results of the European Group for Bone Marrow Transplantation. Stem Cells 1995; 13(supplement 2): 140–146. 85. Shaughnessy Jr J, Barlogie B, McCoy JPJ et al. Early relapse after total Therapy II for multiple myeloma is significantly associated with cytogenetic abnormalities of chromosome 13 (CA13) but not interphase FISH-del 13 or plasma cell labeling index (PCLI). Blood 2001; 98(supplement): 734a. 86. Gertz MA, Witzig TE, Pineda AA et al. Monoclonal plasma cells in the blood stem cell harvest from patients with multiple myeloma are associated with shortened relapse-free survival after transplantation. Bone Marrow Transplant 1997; 19: 337–342. 87. Rajkumar SV, Fonseca R, Dispenzieri A et al. Effect of complete response on outcome following autologous stem cell transplantation for myeloma. Bone Marrow Transplant 2000; 26: 979–983. 88. Barlogie B, Jagannath S, Desikan KR et al. Total therapy with tandem transplants for newly diagnosed multiple myeloma. Blood 1999; 93: 55–65. 89. Alexanian R, Weber D, Giralt S et al. Impact of complete remission with intensive therapy in patients with responsive multiple myeloma. Bone Marrow Transplant 2001; 27: 1037–1043. 90. Sarasquete ME, Garcı´a-Sanz R, Balanzategui A et al, Minimal residual disease studies in multiple myeloma patients achieving complete remission after autologous peripheral blood stem cell transplantation (APBSCT) by RQ-PCR 46th Annual Meeting and Exposition of the American Society of Hematology, San Diego, CA, December 4–7 2004. Abstract #2418. 91. Bakkus MH, Bouko Y, Samson D et al. Post-transplantation tumour load in bone marrow, as assessed by quantitative ASO-PCR, is a prognostic parameter in multiple myeloma. Br J Haematol 2004; 126: 665– 674 (89). 92. Fenk R, Ak M, Kobbe G et al. Levels of minimal residual disease detected by quantitative molecular monitoring herald relapse in patients with multiple myeloma. Haematologica 2004; 89(5): 557–566 (90). 93. Mateo G, Orfao A, Montalba´n MA et al. Immunophenotypic investigation of minimal residual disease in multiple myeloma. Hematol J 2003; 4(supplement 1): S48 (91). 94. Fassas AB, Van Rhee F & Tricot G. Predicting long-term survival in multiple myeloma patients following autotransplants. Leuk Lymphoma 2003; 44: 749–758. 95. Bensinger WI, Buckner CD, Anasetti C et al. Allogeneic marrow transplantation for multiple myeloma: an analysis of risk factors on outcome. Blood 1996; 88: 2787–2793. 96. Gahrton G, Tura S, Ljungman P et al. Prognostic factors in allogeneic bone marrow transplantation for multiple myeloma. J Clin Oncol 1995; 13: 1312–1322.

Prognostic features of multiple myeloma 583 97. Kroger N, Schilling G, Einsele H et al. Deletion of chromosome band 13q14 as detected by fluorescence in situ hybridization is a prognostic factor in patients with multiple myeloma who are receiving allogeneic dose-reduced stem cell transplantation. Blood 2004; 103(11): 4056–4061. Epub 2004 Feb 24. 98. Perez-Simon JA, Martino R, Alegre A et al. Chronic but not acute graft-versus-host disease improves outcome in multiple myeloma patients after non-myeloablative allogeneic transplantation. Br J Haematol 2003; 121(1): 104–108. *99. Greipp Ph, San Miguel JF, Fonsea R et al. On behalf of the International Myeloma Working Group plus Cancer Research and Biostatistics (CRAB). Development of an International Prognostic Index (IPI) for Myeloma: report of the International Myeloma Working Group at the IXth International Workshop on Multiple Myeloma, Salamanca, Spain. May 23–27. The Hematology Journal 2003; 4(supplement 1): S42.