Gene expression profiling advances in multiple myeloma

A state-of-the-art test assesses prognosis and helps individualize therapy.

Multiple myeloma is a fatal disease, and long-term remissions have only recently become possible. The median age of onset is 65 years, and the incidence in the United States is 20,000 per year.

Treatment for multiple myeloma has evolved from melphalan and prednisone, which were of marginal effectiveness, to induction therapy with vincristine, doxorubicin and dexamethasone; in 1999, thalidomide was added to existing regimens, and in 2006, bortezomib (Velcade, Millennium Pharmaceuticals) was approved for treatment. Thalidomide (Thalomid, Celgene) and bortezomib are now considered cornerstones of therapy in addition to stem cell transplant after bone marrow ablation.

According to Kyle and colleagues, the criteria to start therapy includes a serum and/or urine monoclonal paraprotein; greater than 10% monoclonal plasma cells in the bone marrow or biopsy-proven plasmacytoma; and one or more myeloma organ dysfunctions. These include:

• Hypercalcemia (serum calcium >10.5 mg/dL)
• Renal insufficiency (creatinine greater than 2 mg/dL)
• Anemia (hemoglobin, 10 g/dL or > 2 g below normal
• Bone disease

Age-adjusted therapy generally focuses on autologous transplantation for patients aged 25 to 64 years; full-dose chemotherapy for patients aged 65 to 74 years; and reduced-dose therapy for patients older than 75 years. Despite these important advances in available treatment options, using currently available genetic classification systems and diagnostic testing (eg, cytogenetics and fluorescence in situ hybridization testing [FISH]), it is not possible to accurately predict whether a specific patient will respond to current treatments or what the duration of that response will be.

Furthermore, it is impossible to use morphology alone to classify multiple myeloma, and physicians have used clinical criteria such as osteolytic bone lesions, plasma cell burden, anemia (Durie-Salmon classification), the plasma cell labeling index and other chemical parameters (eg, beta-2 microglobulin level) to plan therapies.

Ronald A. Sacher

These principles are somewhat empiric. Moreover, cytogenetics and FISH analysis are meant to further stratify therapy and achieve the greatest potential for success with the least toxic outcome. A comprehensive cytogenetics panel requires the determination of the presence of several abnormal chromosomes abnormalities, including t(4:14), t(14:16) and 17 p13 deletion; t(11:14); chromosome 13 deletion; chromosome 1 abnormalities; and hyperdiploidy.

Limitations of cytogenetics and FISH

Histopathologic, immunologic and chemical criteria are augmented by these two chromosomal tests (cytogenetics and FISH) for the purposes of prognostication and to develop more appropriate therapeutic strategies. These two low-resolution tests provide complementary information. Metaphase cytogenetics is a survey of chromosomal structural and numeric abnormalities; interphase FISH augments cytogenetics and uses fluorescently labeled nucleic acid probes complementary to specific sequences on a chromosome.

With cytogenetic testing, qualitative evaluation of the chromosomal pairs is made to determine whether one or more chromosomal translocations exist. It does not, however, provide useful information on many submicroscopic chromosomal lesions. Moreover, since even in culture multiple myeloma plasma cells retain a low proliferative rate, approximately 80% of karyotype attempts are uninformative (ie, falsely normal).

Interphase FISH detects chromosomal abnormalities that cannot be examined by standard chromosomal analysis or when mitotic cells are not available for chromosomal analysis. Briefly, FISH involves examining denatured interphase chromosomes on a slide using fluorescently labeled DNA probes hybridized to chromosomes and analyzed by fluorescent microscopy. There are a number of different types of FISH probes, including unique sequence probes, whole chromosome painting probes, repetitive probes, probes detecting gene translocations and fusions, as well as other types of chromosomal rearrangements.

FISH can obtain information on specific chromosomes in virtually 100% of cases. This technique can determine whether a specific portion (100kb) of the chromosome is present (gain/amplification), absent (deletion) or fused as a result of translocation. However, FISH is a low resolution test resulting in an incomplete assessment of chromosomal abnormalities (eg, it will miss DNA abnormalities not included in probe sequences).

FISH testing for multiple myeloma is directed at identifying high-risk patients and includes probes for chromosomal regions -13/del 13q, del 17p13.1 (p53), t(4;14), t(11;14), and t(14;16). Thus, this testing is limited and requires two probes for each region. FISH is also used to examine additional chromosomal regions and abnormalities, such as hyperdiploidy (chromosomes 1,5,3,9); 1p deletion/1q21 amplification; t(6;14); t(14;20) or other translocations involving 14q32 using additional probe sets.

The inability to translate better treatments into uniformly better outcomes can be traced to tumor heterogeneity at the genetic/molecular level that cannot be identified using cytogenetic or FISH testing. Thus, there is a critical need to adopt genetic risk stratification methods at the time of initial diagnosis to define and discriminate patients at high risk for early relapse from those at low risk for relapse and to treat patients accordingly.

Armed with such a stratification algorithm, it is anticipated that physicians will have the ability to individualize treatment options, improve therapeutic efficacy and clinical outcomes, minimize adverse effects, perform fewer diagnostic tests, decrease unnecessary treatments and reduce the clinical and financial burden to health care systems and individual patients.

Genetic test for multiple myeloma

Recently, researchers at the University of Arkansas for Medical Sciences (UAMS) have developed a genomic profile test for patients with multiple myeloma, which was licensed for use by Myeloma Health and its parent company, Signal Genetics. The test, called MyPRS, is a gene expression profile that isolates malignant plasma cells from the bone marrow of patients with multiple myeloma and extracts their DNA.

Through the use of state-of-the-art microarray technology and application of validated software, the test provides dynamic information of a select group of genes that are actively being transcribed because it quantifies RNA in purified tumor cells. Importantly, this test is a significant independent variable in multivariate analyses in both UAMS and other public datasets.

The MyPRS test genes were selected using an unbiased genome-wide approach in which the entire repertoire of more than 22,000 genes in the human genome were evaluated. In addition, the 70 most prognostic and relevant genes specific to multiple myeloma, which were shown to best correlate with the likelihood of early relapse, were isolated for the MyPRS profile. These genes affect overlapping steps known to be important for relapse, including cell cycle regulation, angiogenesis, cell adhesion, cell migration, proliferation and signal transduction.

The MyPRS test was developed using more than 10 years of outcome data from a population of more than 1,000 newly diagnosed multiple myeloma patients treated with clinical regimens developed at the UAMS, and validated in other treatment regimens both at UAMS and across the globe. The resulting MyPRS 70-gene profile accurately predicted which patients were at high risk or low risk for early relapse defined as follows: Absence of a high-risk score identifies a favorable subset of patients with a 5-year continuous complete remission of greater than 60% as opposed to a 3-year rate of less than 20% in those with a high-risk score.

This test will help identify patients who will benefit from more intense clinical management, including more aggressive or alternative therapeutic modalities. Compared with standard risk assessment factors and the presence of abnormal metaphase or interphase genetics, the MyPRS test significantly reduces the number of patients traditionally classified with a poor prognosis, or false positives, while identifying those patients who may be at increased risk of relapse despite their standard clinicopathologic and genetic findings (ie, high risk by MyPRS, but International Staging System stage I). Because the MyPRS test uses gene expression profile (GEP) to analyze the gene expression signature of the multiple myeloma itself, it allows physicians to personalize treatment and, in so doing, improves quality of life for more patients. Further, this method of patient stratification enables physicians to more optimally utilize therapeutic and financial resources associated with treating this patient population.

Prognostic risk signature

The specific output of the MyPRS test is the prognostic risk signature. This generated prognostic risk signature (the MyPRS score) explains the residual risk of relapsing with active multiple myeloma in a specific period of time. Once expression values for each of the 54,675 probes have been confirmed as being within acceptable assay limits for a patient sample, the expression for the 70 risk-related genes is normalized relative to the average reference gene values. The expression values are then used to generate the risk score. The score for each specimen is then reported and interpreted by the treating physician by reference to the continuous curve of risk values determined from several published, clinical validation studies. Based on the published medical evidence, the test should no longer be considered as solely a research tool and should be used as part of routine patient management. These data may justify the test supplanting cytogenetic and FISH testing.

Modern array testing makes it possible to completely characterize a tumor at the molecular level and account for the clinical heterogeneity of multiple myeloma on a patient-by-patient basis. It also enables doctors to classify a patient’s multiple myeloma based on its genomic expression profile and not just a few chromosomal abnormalities, many of which have lost their prognostic relevance in recent years (ie, del 13 and t[4;14]). Besides providing detailed genetic data, the MyPRS gene expression profile includes a risk score calculated using a complex computer algorithm that provides an accurate prediction of the patients expected survival given state-of-the art treatments.

In terms of clinical utility, the data suggest that MyPRS gene expression profile testing should be performed in all patients diagnosed with multiple myeloma. In the asymptomatic stage, the altered expression levels of specific genes involved in bone destruction (eg, DKK1) or cellular proliferation can forecast the probability of progression to active myeloma. In active symptomatic myeloma, the test can provide prognostication of survival probabilities of 10+ years vs. 2 years that cannot be predicted by any other test alone or in combination.

A second critical clinical event requiring an additional GEP analysis is at the time of relapse, which UAMS research has shown has a high probability of reflecting an evolution to a high-risk MyPRS score in most patients. A conversion to high risk at relapse significantly reduces post–relapse survival. All newly diagnosed multiple myeloma patients can now get a GEP analysis at baseline before initiation of treatment, stratified according to prognostic profile, and thus therapy modifications can be incorporated into the clinical management of the patient

Approximately 30% of low-risk patients will relapse within the first 5 years and will need to be reassessed with the GEP test to determine whether the tumor has undergone genetic evolution to high-risk, thus providing the rationale for modified salvage regimens.

In the age of personalized medicine, these newer diagnostic techniques offer unique advantages to help the clinician better manage their patients. In the case of the MyPRS test, clinicians and patients can both benefit from the fact that the test can prospectively identify the approximately 80% of newly diagnosed disease that will have an excellent prognosis with the use of current treatments, and also those whose disease will require more aggressive novel approaches. Advances in the understanding of the molecular profile of the disease will also likely further the development of new therapeutics for rare diseases.

Ronald A. Sacher, MD, FRCPC, is a professor of internal medicine and pathology and director of Hoxworth Blood Center at the University of Cincinnati Academic Health Center. Dr. Sacher is also a HemOnc Today Section Editor.

For more information:

• Kyle RA. Leukemia. 2009;23:3-9.
• Shaughnessy. Blood. 2007;109:2276-2284.
• Stewart AK. Blood. 2010;116:674-675.