The accumulation of abnormal plasma cells within the bone marrow is the hallmark of multiple myeloma. Although these cells are cancerous, they resemble normal plasma cells in several ways. For example, both normal and myeloma plasma cells are capable of producing immunoglobulins or antibodies. In addition, normal plasma cells are fully mature (or “terminally differentiated”), meaning that they have lost the ability to divide and form new cells. Myeloma plasma cells also appear to be incapable of replicating; thus, it has been unclear how they are formed. Our research focuses on studying the originating cell in myeloma (the myeloma stem cell) and developing treatments that target this cell and inhibit the production of new tumor cells.
Cancer Stem Cells
Most cells in the body, such as those that make up the outermost layer of the skin, red cells, and neutrophils in the blood, and the neurons within the brain are terminally differentiated like normal plasma cells. As these cells are lost or used up they need to be replaced by new cells, but terminally differentiated cells cannot normally divide. Instead, they are formed by less mature cells known as stem cells. Stem cells are distinct from differentiated cells because they are capable of dividing many times. When a stem cell divides, two new cells are formed. One of the daughter cells goes on to produce differentiated cells by dividing a specific number of times, and with each division the resulting cells gradually mature and expand in number. Importantly, these maturing cells also lose their ability to divide an unlimited number of times. The other cell remains an exact copy of the original stem cell. This process is known as self-renewal, and it ensures that an adequate number of stem cells will be available to form new differentiated cells as needed. Thus, the vast majority of cells that make up most tissues and organs in the body are terminally differentiated and cannot replicate long-term. In contrast, stem cells are rare but have the capacity to divide an unlimited number of times.
Like normal tissues, human cancers also appear to consist of both differentiated cells and stem cells. Several groups have examined the growth of human cancers, either in the laboratory or in animals, and have found that the majority of cells within an individual tumor lack the ability to form new cancer cells. These results may be explained by two possibilities.1 The first is that all the cancer cells have an equal ability to form new tumors, but only a minority is able to do so at any point in time. The second is that the cells are not identical, and the ability to divide and grow is limited to only a few specialized cells (i.e., cancer stem cells). For many years, there was speculation that cancer stem cells existed, but they were not identified until the early 1990s when they were isolated in chronic myeloid leukemia and acute myeloid leukemia.2-3 Because of the rarity of stem cells, they were not found in any other cancer until just a few years ago when they were identified in breast cancer and brain tumors.4-6
Studies examining cancer stem cells in a variety of unrelated tumor types have provided several important gener-alities regarding their behavior. First, the methods used to study normal stem cells that form blood cells (hematopoietic stem cells) and nerve cells (neural stem cells) can also be used to isolate myeloid leukemia and brain cancer stem cells. Thus, cancer stem cells highly resemble their normal counterparts. Furthermore, when cancer stem cells are propagated in the laboratory or injected into mice, they form cancers consisting of mature cells that are identical to the original tumor. These findings suggest that cancer stem cells can also undergo some degree of differentiation in addition to self-renewal. Therefore, cancer stem cells appear to be very similar to normal stem cells.
Multiple Myeloma Stem Cells
In multiple myeloma, the accumulation of abnormal plasma cells within the bone marrow is responsible for the main symptoms associated with the disease, namely bone disease, kidney failure, anemia, and a susceptibility to infections. Although the myeloma plasma cells are abnormal, several studies have suggested that they are terminally differentiated like normal plasma cells and rarely divide. One study found that the majority of the plasma cells located within the bone marrow of myeloma patients did not undergo cell division.7 Likewise, several other studies examining the growth of myeloma patient samples in the laboratory demonstrated that only 1 in 1,000 to 1 in 100,000 were able to form new plasma cells.8-9 It is well known that plasma cells are normally produced by B cells which play an important role in the immune sys-tem. Therefore, several studies examined whether B cells related to the myeloma plasma cells could be found within patients. In these studies, B cells with the same DNA sequences used to produce the abnormal immunoglobulin by myeloma plasma cells were found within the blood and bone marrow of myeloma patients.10-12 Furthermore, mice injected with B cells from a patient with myeloma developed the disease.13 Taken together, these studies sug-gest that myeloma plasma cells may be formed by B cells rather than plasma cells.
We studied whether plasma cells or B cells represent the cancer stem cell in myeloma, and initially developed a special assay based on previous work that allowed us to grow patient myeloma samples in the laboratory.14-15 In this assay, bone marrow cells are collected from myeloma patients, and then placed in a specialized culture medium. This medium is very viscous so that cells cannot move freely, and as cells divide they form clusters, termed colonies, that can be easily identified and counted with a microsope. In order to determine whether myeloma plasma cells were able to form colonies, we specifically isolated plasma cells using a specific antibody which recognizes a protein called syndecan-1, or CD138, that is expressed only by plasma cells in the normal immune system. Plasma cells that expressed CD138 (CD138+) did not form colonies, whereas cells that lacked CD138 (CD138neg) could form myeloma colonies in our assay. We also injected CD138+ or CD138neg cells into mice and found that only the mice receiving CD138neg cells developed myeloma. Taken together, these results suggested that CD138+ myeloma plasma cells lack the ability to divide and form new cells, but that the stem cell was CD138neg. Since cancer stem cells resemble their normal counterparts in myeloid leukemias and normal plasma cells come from B cells, we next examined whether the cells able to form colonies in our assay resembled B cells. We took CD138neg bone marrow cells, removed B cells using specific antibodies against B cell proteins (CD45, CD19, CD20, and CD22), then evaluated colony formation in our laboratory assay. The removal of cells expressing any of these proteins from the starting pool of CD138neg cells markedly decreased colony growth, indicating that myeloma stem cells were likely B cells.
Because of the difficulties obtaining repeated bone marrow samples, we also examined a number of myeloma cell lines. These were originally derived from myeloma patients, but unlike most patient samples, these cell lines can be easily and indefinitely propagated in the laboratory. Similar to myeloma patient samples, we found that the majority of cells within these cell lines were CD138+ plasma cells. However, a small population of CD138neg cells that expressed B cell, rather than plasma cell, markers could be reliably found. We also found that these CD138neg cells had much greater potential to form colonies than the cor-responding CD138+ cells. These studies suggested that we could use these cell lines in addition to patient samples to study myeloma stem cells.
Strategies to Inhibit Multiple Myeloma Stem Cells
Our studies demonstrating that myeloma stem cells are B cells suggested that therapies that directly target B cells could inhibit their ability to form new cells. In addition to myeloma, B cells are the origin of several other cancers, such as certain lymphomas and leukemias.16 Since a number of therapies specifically designed to inhibit cancerous B cells are currently used to treat patients with lymphoma, we examined whether one such drug, rituximab, could affect the growth of myeloma stem cells. Rituximab is a monoclonal antibody against CD20, a protein expressed on the surface of B cells and present on myeloma stem cells. We treated CD138neg cells from patient samples with rituximab, then evaluated them for colony formation. We found that after only 24 hours, rituximab was able to inhibit the growth of myeloma colonies in our assay by 65%.15
Testing This Approach in a Clinical Trial
If myeloma were analogous to a dandelion, the visible part of the weed would represent the plasma cells, whereas the root would symbolize the stem cells. Many available thera-pies significantly decrease the burden of plasma cells and improve symptoms in patients with myeloma, but relapse following treatment suggests that they are inactive against myeloma stem cells. This might be similar to cutting a dandelion off at ground level and eliminating the visible portion of the weed, but leaving the root untouched would allow it to grow back. Conversely, treatments like rituximab that selectively attack myeloma stem cells would not appear to work immediately since the plasma cells are not targeted. However, the inhibition of myeloma stem cells may lead to long-term remissions. An ideal strategy may be to combine therapies that target both myeloma plasma cells and stem cells. This would allow rapid improvement in symptoms caused by the plasma cells as well prevent the growth of new tumor cells.
Based on our laboratory studies, we have begun a phase II clinical trial to study the activity of rituximab against cancer stem cells in patients with myeloma. One of the challenges in studying agents that are designed to target cancer stem cells, like rituximab, is determining how best to measure the effectiveness of the treatment. Cancer stem cells represent a very small percentage of the tumor cells that are present in patients with cancer. Since standard response criteria measure changes in tumor bulk, they are not likely to reflect changes in the stem cell population, even if they are significant. That is, serum and urine pro-tein electrophoreses and bone marrow aspirates which are standard measures to assess response in myeloma are likely not helpful in measuring the effectiveness of rituximab against myeloma stem cells.
If these are the sole criteria used, an approach that is effective against stem cells might be discarded prematurely because it appeared to be ineffective. This may explain why a prior trial of rituximab in myeloma concluded that it was not effective in treating myeloma.17 In this trial, 19 patients with myeloma were treated with rituximab for 4 weeks, and responses, using standard response criteria, were measured at 3 months. Six of the 19 patients had responses (one partial remission and five with stable disease) with at least one patient with a response extending beyond two years. Thus, if the myeloma stem cells could be measured directly, it is possible that response rates might be higher. An alternative approach is to measure survival, and in particular progression-free survival, in looking at the effective-ness of treatment directed at rare populations of cells, as this may be a better reflection of efficacy.
Rituximab is being used to treat the myeloma stem cells. Although rituximab is active in vitro against the myeloma stem cell, it does not appear to have much activity against the mature plasma cells that make up the bulk of the tumor.15 Thus, patients in this trial are being treated with a combination of cyclophosphamide, a drug with known activity against plasma cells, and rituximab in an effort to target the myeloma stem cell.
Cyclophosphamide is a chemotherapeutic agent that works by crosslinking DNA and interfering with cell division. It is a drug with known activity against plasma cells. Thus, it is being given in high doses to reduce the number of plasma cells in patients with myeloma and facilitate rituximab’s ability to reach the myeloma stem cells. Cyclophosphamide has a unique pharmacology. It is a prodrug and must be metabolized in the body before it is active. Further, it does not kill all cells that are growing and dividing. Cells with high concentrations of the enzyme aldehyde dehydrogenase are resistant to cyclophosphamide. One population of cells that are not affected by cyclophosphamide are normal hematopoeitic progenitors, as they have a high concentration of this enzyme. Thus, even though high doses are given, the normal hemato-poietic progenitors are unaffected by this and thus, fol-lowing a relatively limited duration of low blood counts, all patients recover their normal hemoglobin, neutrophils, and platelets without the need for a stem cell transplant.
The primary objective of the trial is to measure progression-free survival one year after treatment. While standard responses will also be measured, we believe that assessing survival will be a better reflection of activity against the myeloma stem cell than the determination of partial and complete remissions. Equally important in this trial are the secondary objectives of using our laboratory assay to measure myeloma stem cells throughout the course of treatment and follow-up. At specified points in the treat-ment schema, clonogenic assays will be done to measure the effects of treatment on the myeloma stem cells.
We will then use statistical models to correlate the results of the laboratory assays with the clinical outcomes and in so doing, hope to develop a model for measuring disease response with in vitro assays prior to the determination of a survival benefit. If this can be done, it will enable us to speed the process of developing new therapies and develop more efficient ways to assess the efficacy of novel treatments.
As curing myeloma is the ultimate goal of everyone’s efforts to understand and treat myeloma, we believe that novel approaches such as this offer the potential to better target the myeloma stem cell, giving us insight into better treatments for this disease.
Who Is Eligible?
Patients are eligible to participate if they have high-risk myeloma in first remission, either partial or complete. High-risk disease in this trial is defined as chromosome 13 deletion by FISH or cytogenetics, or beta-2 microglobulin > 5 mg/dL. Patients are also eligible: a) if they have relapsed myeloma that is responding to treatment, or b) immediately after their initial therapy if the treatment led to less than a partial remission.
Patients will undergo all screening tests to ensure that they meet the eligibility requirements and have adequate cardiac and pulmonary function to proceed. Once con-firmed, patients will receive two doses of rituximab four days apart, followed by four daily doses of cyclophosphamide the following week. Approximately three weeks after completion of the cyclophosphamide, when blood counts have recovered, patients will be given rituximab once a week for four weeks. Rituximab will also be given on a maintenance schedule every 3 months beginning with month 3 from the start of treatment.
Safety, Tolerability, and Monitoring
Both rituximab and cyclophosphamide have been given to thousands of patients with hemtologic malignancies and their safety and tolerability are well known. Rituximab has been used predominantly in the treatment of lymphomas, either alone or in combination with chemotherapy, including cyclophosphamide. Cyclophosphamide is an active agent in the treatment of myeloma both in combination with corticosteroids and as a part of the chemotherapeutic preparative regimens used in bone marrow transplantation. Patients will be monitored closely with serial assessments both of tolerance and of disease status by means of serum and urine protein electrophoresis and through bone marrow analyses. Secondary studies will be done in the laboratory to measure the effectiveness of treatment using measurements of myeloma colony formation and assessments of the mechanism of rituximab’s action.
The results of this trial and our laboratory studies will allow us to examine the efficacy of combining agents with activity against both myeloma stem cells and mature plasma cells. We expect that the results of both these endeavors will allow us to design and develop future clinical trials with the ultimate goal of producing long-term remissions.
1. Reya T, Morrison SJ, Clarke MF and Weissman IL. 2001. Stem cells, cancer, and cancer stem cells. Nature. 414: 105-111.
2. Bedi A, Zehnbauer BA, Collector MI, Barber JP, Zicha MS, Sharkis SJ and Jones RJ. 1993. BCR-ABL gene rearrange-ment and expression of primitive hematopoietic progenitors in chronic myeloid leukemia. Blood. 81: 2898-2902.
3. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang TC-CJ, Minden M, Paterson B, Caligiuri MA and Dick JE. 1994. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 367: 645-648.
4. Al Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ and Clarke MF. 2003. Prospective identification of tumori-genic breast cancer cells. Proc Natl Acad Sci U S A. 100: 3983-3988.
5. Hemmati HD, Nakano I, Lazareff JA, Masterman-Smith M, Geschwind DH, Bronner-Fraser M and Kornblum HI. 2003. Cancerous stem cells can arise from pediatric brain tumors. PNAS. 100: 15178-15183.
6. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, Henkelman RM, Cusimano MD and Dirks PB. 2004. Identification of human brain tumour initiating cells. Nature. 432: 396-401.
7. Drewinko B, Alexanian R, Boyer H, Barlogie B and Rubinow SI. 1981. The growth fraction of human myeloma cells. Blood. 57: 333-338.
8. Hamburger A and Salmon SE. 1977. Primary bioassay of human myeloma stem cells. J Clin Invest. 60: 846-854.
9. Takahashi T, Lim B, Jamal N, Tritchler D, Lockwood G, McKinney S, Bergsagel DE and Messner HA. 1985. Colony growth and self renewal of plasma cell precursors in multiple myeloma. J Clin Oncol. 3: 1613-1623.
10. Bakkus MH, Van R, I, Van Camp B and Thielemans K. 1994. Evidence that the clonogenic cell in multiple myeloma originates from a pre-switched but somatically mutated B cell. Br J Haematol. 87: 68-74.
11. Billadeau D, Ahmann G, Greipp P and Van Ness B. 1993. The bone marrow of multiple myeloma patients contains B cell populations at different stages of differentiation that are clonally related to the malignant plasma cell. J Exp Med. 178: 1023-1031.
12. Jensen GS, Mant MJ and Pilarski LM. 1992. Sequential maturation stages of monoclonal B lineage cells from blood, spleen, lymph node, and bone marrow from a terminal myelo-ma patient. Am J Hematol. 41: 199-208.
13. Pilarski LM, Seeberger K, Coupland RW, Eshpeter A, Keats JJ, Taylor BJ and Belch AR. 2002. Leukemic B cells clonally identical to myeloma plasma cells are myelomagenic in NOD/SCID mice. Exp Hematol. 30: 221-228.
14. Matsui W, Huff CA, Vala M, Barber J, Smith BD and Jones RJ. 2003. Anti-tumour activity of interferon-alpha in multiple myeloma: role of interleukin 6 and tumor cell differentiation. Br J Haematol. 121: 251-258.
15. Matsui W, Huff CA, Wang Q, Malehorn MT, Barber J, Tanhehco Y, Smith BD, Civin CI and Jones RJ. 2004. Characterization of clonogenic multiple myeloma cells. Blood. 103: 2332-2336.
16. Kuppers R, Klein U, Hansmann ML and Rajewsky K. 1999. Cellular origin of human B-cell lymphomas. N Engl J Med. 341: 1520-1529.
17. Treon SP, Pilarski LM, Belch AR, Kelliher A, Preffer FI, Shima Y, Mitsiades CS, Mitsiades NS, Szczepek AJ, Ellman L, Harmon D, Grossbard ML and Anderson KC. 2002. CD20-directed serotherapy in patients with multiple myeloma: biologic considerations and therapeutic applications. J Immunother. 25: 72-81.
NOTE: This clinical trial is open at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins and is actively accruing patients. If you would like more information or are interested in participating, please contact either Carol Ann Huff, MD, at 443-287-7104 or Kathryn Rogers, RN, at 410-614-1766.