Original article
`
`Annals of Oncology 12: 987-990. 2001.
`© 2001 Kluwer Academic Publishers. Printed in the Netherlands.
`
`Thalidomide in multiple myeloma, myelodysplastic syndromes and
`histiocytosis. Analysis of clinical results and of surrogate angiogenesis
`markers
`
`F. Bertolini,1* W. Mingrone,2* A. Alietti,1 P. F. Ferrucci,1 E. Cocorocchio,1 F. Peccatori,1
`S. Cineri,1 P. Mancuso,1 C. Corsini,1 A. Burlini,1 E. Zucca2 & G. Martinelli'
`1 Division of Hematology-Oncology, European Institute of Oncology, Milan, Italy, 2Oncology Institute of Southern Switzerland. Locarno.
`Switzerland
`
`Summary
`
`Background: Thalidomide, as a single agent, has been recently
`found to induce a clinical response in one third of refractory
`or relapsed myeloma patients. Although it has been reported
`that thalidomide significantly inhibits angiogenesis, it is still
`unclear whether its clinical effect is mediated, at least in part,
`by its anti-angiogenic properties.
`Patients and methods: We evaluated thalidomide as a single
`agent in myeloma, myelodysplastic syndromes (MDS) and
`histiocytosis, i.e. hematological diseases characterized by in-
`creased angiogenesis, and measured prospectively a number of
`surrogate angiogenesis markers.
`Results: Clinical responses were observed in 7 of 17 mye-
`
`loma and 2 of 5 MDS patients. The histiocytosis patient had a
`partial response. At the time of the best clinical response,
`plasma levels of angiogenic growth factors, vascular endothe-
`lial growth factor (VEGF) and basic-fibroblast growth factor
`(b-FGF), were significantly decreased, and flow cytometry
`indicated a decrease of activated endothelial cells in the bone
`marrow of responding MDS patients.
`Conclusions: These observations confirm thalidomide effi-
`cacy in myeloma, suggest a possible use in MDS and histio-
`cytosis and may contribute to the prediction of clinical response
`and to understanding the mechanism of thalidomide's action.
`
`Key words: angiogenesis, histiocytosis. myelodysplastic syn-
`dromes, myeloma, thalidomide
`
`Introduction
`
`Thalidomide, developed in the 1950s as a sedative-
`hypnotic, was withdrawn in the 1960s after reports of
`teratogenicity associated with its use. Thereafter, thali-
`domide has been used for leprosy, discoid lupus ery-
`thematosis, aphthous ulcers in HIV syndromes and
`Behcet's disease [1]. D'Amato et al. [2] have reported
`that, in addition to already known actions (including
`effects on the expression of adhesion molecules, of
`cytokines such as TNF-oc and IL-10, and modulation
`of cell-mediated immunity), thalidomide significantly
`inhibits angiogenesis. Vacca et al. [3] described relevant
`neovascularization in the bone marrow (BM) of mye-
`loma patients, and Singhal et al. [4] reported that thali-
`domide, as a single agent, was able to induce a marked
`and durable response in approximately one third of 84
`chemotherapy-refractory myeloma patients. Although
`two more reports have confirmed the clinical efficacy of
`thalidomide in small series of myeloma patients [5-6],
`it is still unclear whether this effect is mediated, at least
`in part, by its anti-angiogenic properties.
`We have recently reported that angiogenesis is in-
`
`* These authors contributed equally to this work.
`
`creased in myelodysplastic syndromes (MDS), and that
`in this disease BM neovascularization is intermediate
`between healthy controls and acute myeloid leukemia
`[7]. Thus, we decided to evaluate thalidomide as a single
`agent in relapsed myeloma and MDS patients, and to
`measure a number of surrogate markers of angiogenesis
`to gain insight into the biological activity of this drug.
`Here we report an intention-to-treat analysis of 17 mye-
`loma and five MDS patients enrolled in this study and of
`a patient suffering from histiocytosis, another disease
`characterized by increased angiogenesis.
`
`Study design
`
`The median age of the patients was 66 years (range 55-
`80). Of the myeloma patients enrolled, 3 of 17 relapsed
`after tandem high-dose chemotherapy, and 14 of 17
`relapsed after (or were refractory to) at least two lines
`of conventional chemotherapy. MDS patients (two RA,
`three RAEB) failed to respond to cytokine-based thera-
`pies and were treated with supportive care, including red
`cell and platelet transfusions. The histiocytosis patient
`
`DR. REDDY’S LABS., INC. EX. 1046 PAGE 1
`
`

`

`988
`
`was in relapse after a previous velbe-steroid treatment.
`Thalidomide was administered orally at a starting dose
`of 100 mg/day. In the absence of side effects, the dose
`was increased every 2-4 weeks up to 3-400 mg/day.
`Singhal et al. [4] administered thalidomide in a single
`dose in the evening. Similarly to Juliusson et al. [6], and
`taking into account the 6 hours half-life of the drug [8],
`our patients were requested to fractionate the dose in at
`least two daily administrations. Patients were evaluated
`monthly for clinical response and every three months for
`plasma levels of angiogenic growth factors, vascular
`endothelial growth factor (VEGF) and basic-fibroblast
`growth factor (b-FGF), as we previously described [9].
`In MDS patients, the frequency of resting and activated
`BM endothelial cells was evaluated by flow cytometry
`following a procedure that we previously validated in
`preclinical models of human hematopoietic malignan-
`cies [10], and in a clinical study enrolling lymphoma
`and breast cancer patients [11]. All procedures were in
`accordance with the ethical standards of the responsible
`committees on human experimentation and with the
`Helsinki Declaration of the World Medical Association
`
`Results
`
`Side effects were observed in 10 of 23 patients. Three
`patients had constipation, two had Gl-2 neuropathy,
`two had dizziness, one had hypotension. In two patients
`the drug was discontinued for general or gastric intoler-
`ance. Patients without side effects were able to tolerate a
`higher dose of thalidomide. In fact, 7 of 23 patients
`failed to escalate the dose up to 3-400 mg/day because
`of the side effects. On the other hand, split daytime
`dosage did not increase somnolence, fatigue or other
`side effects. One patient reduced the dose to 50 mg/day
`after achieving a clinical response. Overall, the median
`thalidomide dose per day was 220 mg/day (range 50-
`400).
`Of the 17 myeloma patients enrolled, five (29%) had a
`> 50% decrease of the monoclonal Ig, two had a 26%-
`50% reduction, five had a l%-25% reduction, three had
`progressive disease (PD), two discontinued the drug
`within the first 30 days. In responding patients, median
`time to response was 36 days. In this limited group of
`patients, a clear evaluation of dose-response relation-
`ship was not possible. We observed durable clinical
`responses in some patients who achieved a response at
`3-400 mg/day and reduced the drug dosage to 1-200
`mg/day to avoid side effects. Apart from two patients
`who discontinued the drug for side effects, other patients
`discontinued because of progressive disease. Interest-
`ingly, the patient who achieved the best clinical response
`(normalization of Ig levels with 50 mg/day thalidomide
`for more than one year) was the one with highest base-
`line VEGF and b-FGF levels. Of the seven patients with
`a > 25% decrease in monoclonal Ig, three had PD
`(median time to progression 14 months) after a median
`follow-up of 16 months.
`
`(a)
`
`700 -i
`
`(b)
`
`160 -i
`
`Before
`
`After
`
`Before
`
`After
`
`Figure 1. Plasma levels of (a) VEGF and (b) b-FGF before thalido-
`mide administration and at the time of best clinical response in
`myeloma (solid lines), MDS (small dot lines) and histiocytosis (large
`dot line) patients. P = 0.020 and 0 027 by Wilcoxon-matched pair test
`for VEGF and b-FGF. respectively.
`
`Of the five MDS patients, two had a clinical response
`as defined by a recent consensus report [12]. A 64 year-
`old male RAEB patient became transfusion-independent,
`and his BM blasts were reduced from 20% to 5%. Time
`to progression was four months. A 70 year-old female
`RA patient had increased red cell, white cell and platelet
`count, and transfusion interval increased from 15 to 60
`days. Time to progression was seven months. The 63
`year-old female histiocytosis patient had a significant
`decrease of cutaneous lesions and pain, and time to
`progression was six months.
`As shown in Figure 1, at the time of the best response
`both VEGF and b-FGF were significantly decreased
`when compared to pre-treatment values (P = 0.020 and
`0.027 by Wilcoxon matched pair test for VEGF and
`b-FGF, respectively). As reported in Figure 2, flow
`cytometry indicated a striking decrease of activated
`endothelial cells (CD45-, CD14-, CD31+, CD105+) in
`the BM of responding MDS patients.
`
`Discussion
`
`Our data confirm that thalidomide, as a single agent, is
`active in a significant proportion of refractory myeloma
`
`DR. REDDY’S LABS., INC. EX. 1046 PAGE 2
`
`

`

`989
`
`• •;. '.'- , ~b%-/::
`
`..
`
` -
`
`ooo
`
`o>
`
`FSC-Height
`
`1000
`
`FSC-Height
`
`1000
`
`R1
`
`ooo
`
` -
`
`1 °
`
`CD45 PerCP
`
`10
`
`10
`
`10"
`10
`CD4S PerCP
`
`104
`
`o
`
`CO
`
`10"
`10
`CD105FITC
`
`10
`
`o
`
`0.07%
`
`o .
`10° 101
`1
`10
`CD105FITC
`
`)3 104
`
`<D
`
`X6C
`
`O
`CO
`
`Figure 2. Representative dot plots of the frequency of bone marrow-activated endothelial cells evaluated by flow cytometry in a MDS patient
`before (panels on the left) and after (panels on the right) thalidomide Upper panels show the side and the forward scatter of the bone marrow
`cells. A significant Increase of large blasts is observed in the left panel (before thalidomide). whereas a normal distribution is observed after
`therapy. Middle panels show the gate used to exclude CD45+ hematopoietic cells. Panels on the bottom show the frequency of activated
`endothelial cells (CD45-. CD14-. CD31+, CDI05+). A ten fold reduction of activated endothelial cells and return to normal values was observed
`after thalidomide treatment.
`
`patients, and indicate a promising clinical activity of
`single-agent thalidomide in both MDS and histiocyto-
`sis, two hematological malignancies that (similarly to
`myeloma) are associated with relevant angiogenesis.
`Along this line, we observed that circulating angiogenic
`growth factors VEGF and b-FGF, as well as BM acti-
`vated endothelial cells, are significantly decreased at the
`time of best clinical response. It should be noted that,
`unlike serum VEGF, plasma VEGF is not influenced by
`VEGF released from platelets [9, 13]. Thus, fluctuations
`in the platelet count of myeloma and, particularly, MDS
`patients did not bias our measurements of circulating
`VEGF.
`In myeloma [14] and MDS [7], VEGF and b-FGF are
`involved in paracrine loops generated by stroma, endo-
`thelial and neoplastic cells. Thus, our findings support
`the hypothesis that the action of thalidomide might be
`due (at least in part) to the inhibition of cytokine-signal-
`ling between stroma, endothelial and neoplastic cells
`[15]. Sezer et al. [16] recently reported increased VEGF
`and b-FGF levels in myeloma patients compared to
`
`controls. VEGF and b-FGF were found to decrease in
`patients responding to chemotherapy and not in patients
`who did not achieve a remission. Accordingly, the de-
`crease of VEGF and b-FGF observed in the present
`study in myeloma patients at the time of the best
`response to thalidomide might also reflect a decrease in
`tumor load.
`Another relevant issue is whether high levels of
`VEGF, b-FGF and/or BM-activated endothelial cells
`may predict a clinical response to thalidomide. Although
`our series of patients is too small to draw definitive
`conclusions, it should be noted that myeloma patients
`showing higher VEGF and b-FGF levels had clinical
`responses, and that the two MDS patients with higher
`levels of BM-activated endothelial cells responded to
`thalidomide. Singhal et al. [4] reported a decrease in
`BM microvessel density (MVD) in some myeloma pa-
`tients responding to thalidomide, but differences in MVD
`between responding and non-responding patients were
`not statistically significant. Our flow cytometry assay
`generates quantitative data on the frequency of endo-
`
`DR. REDDY’S LABS., INC. EX. 1046 PAGE 3
`
`

`

`990
`
`thelial cells, while MVD indicates the frequency of
`blood vessels. In animal models of human myeloid and
`lymphoid malignancies, tumor engraftment potential,
`speed of engraftment and the frequency of apoptotic
`tumor cells correlated better with the frequency of endo-
`thelial cells enumerated by flow cytometry than with
`MVD evaluation [10]. We are currently enrolling more
`myeloma and MDS patients and evaluating different
`biological parameters to further elucidate the role of
`surrogate angiogenesis markers as prognostic and/or
`predictive factors.
`
`References
`
`1. Thomas DA. Kantarjian HM. Current role of thalidomide in
`cancer treatment. Curr Opin Oncol 2000. 12: 564-73
`2. D'Amalo RJ. Loughnan MS, Flynn E, Folkman J. Thalidomide is
`an inhibitor of angiogenesis. Proc Natl Acad Sci USA 1994; 91:
`4082-5.
`3. Vacca A. Ribatti D, Roncali L et al. Bone marrow angiogenesis
`and progression in multiple myeloma. Br J Haematol 1994; 87:
`503-8.
`4. Singhal S. Mehta J. Desikan R et al. Antitumor activity of
`thalidomide in refractory multiple myeloma. New Engl J Med
`1999:341: 1565-71
`5. Kneller A. Raanani P. Hardan 1 et al. Therapy with thalidomide
`in refractory multiple myeloma patients-the revival of an old
`drug. Br J Haematol 2000; 108: 391-3.
`Juliusson G. Celsing F. Turesson 1 et al. Frequent good partial
`remissions from
`thalidomide including best response ever in
`patients with advanced refractory and relapsed myeloma. Br J
`Haematol 2000; 109. 89-96.
`7. Pruneri G. Bertolini F. Soligo D et al. Angiogenesis in myelodys-
`plastic syndromes Br J Cancer 1999; 81: 1398-401.
`8. Stirling D. Pharmacology of thalidomide. Sem Hematol 2000. 37
`(Suppl 3): 5-14
`
`6
`
`9. Bertolini F. Paolucci M. Peccatori F et al. Angiogenic growth
`factors and endostatin in non-Hodgkin's lymphoma. Br J Hae-
`matol 1999; 106: 504-9.
`10. Fusetti L, Gobbi A, Rabascio C et al. Human myeloid and
`lymphoid malignancies in the NOD/SCID mouse model: Fre-
`quency of apoptotic cells in solid tumors and efficiency and speed
`of engraftment correlate with VEGF production. Cancer Res
`2000; 60: 2527-34.
`11. Mancuso P, Burlini A, Pruneri G et al. Resting and activated
`endothelial cells are increased in the peripheral blood of cancer
`patients. Blood 2001; 97: 3658-61.
`12. Cheson BD Bennett JM; Kantarjian H et al. Report of an
`international working group to standardize response criteria for
`myelodysplastic syndromes. Blood 2000; 96: 3671-4
`13. Wynendaele W. Derua R. Hoylaerts MF et al. Vascular endo-
`thelial growth factor measured in platelet poor plasma allows
`optimal separation between cancer patients and volunteers: A key
`to study an angiogenic marker in v/vo? Ann Oncol 1999; 10:
`965-71.
`14. Dankbar B, Padro T. Leo R et al. Vascular endothelial growth
`factor and interleukin-6 in paracrine tumor-stromal cell inter-
`actions in multiple myeloma. Blood 2000; 95: 2630-6.
`15 Bertolini F, Mancuso P, Gobbi A. Pruneri G The thin red
`line: Angiogenesis in normal and malignant hematopoiesis. Exp
`Hematol 2000; 28: 993-1000.
`16. Sezer O. Jakob C, Eucker J et al. Serum levels of the angiogenic
`cytokines b-FGF, VEGF and HGF in multiple myeloma. Eur J
`Haematol 2001; 66-83-8.
`
`Received 21 February 2001, accepted 11 April 2001.
`
`Correspondence to:
`F. Bertolini. MD. PhD
`Division of Hematology-Oncology
`European Institute of Oncology
`via Ripamonti 435
`20141 Milan
`Italy
`E-mail: francesco.bertolini@ieo.it
`
`DR. REDDY’S LABS., INC. EX. 1046 PAGE 4
`
`