|
|
| |||||||||||||||||||||||||||||||||||||||
Background Chronic idiopathic thrombocytopenic purpura is a destructive thrombocytopenia caused by an autoantibody. About 80 percent of patients with chronic idiopathic thrombocytopenic purpura have remissions after either corticosteroid therapy or splenectomy. Some patients with resistant disease respond to other agents, but a substantial group are refractory to therapy.
Methods We used combination chemotherapy to treat 10 patients with refractory immune thrombocytopenia. An average of 6.8 (range, 3 to 10) previous therapies, including corticosteroids and splenectomy, had been unsuccessful in these patients. The patients received from three to eight cycles of therapy consisting of cyclophosphamide and prednisone combined with either vincristine (one patient), vincristine and procarbazine (four patients), or etoposide (six patients, including one patient who received four cycles each containing procarbazine and etoposide).
Results Among the 10 patients, 6 had complete responses (platelet count, >180,000 per cubic millimeter); of these, 4 patients had responses that persisted for more than 11, 30, 54, or 126 months, 1 had a relapse 9 months after therapy but had a remission with further therapy and remained in remission for 48 months before dying of an unrelated illness, and another relapsed just before her fifth course of therapy. Two patients had partial responses (platelet count, >50,000 per cubic millimeter); the platelet counts in one remained stable for more than nine months after the end of therapy, and the other patient relapsed. The remaining two patients had no response. Complete responses were associated with a disappearance or marked decrease in the level of platelet-associated autoantibody.
Conclusions Combination chemotherapy is beneficial in some patients in whom immune thrombocytopenia is refractory to corticosteroids and splenectomy.
In 1981, a 29-year-old woman presented to our institution with immune thrombocytopenia. She had been treated with radiation therapy 13 years earlier for stage IIA Hodgkin's disease. She relapsed 3 years later, and over a 1 1/2-year period received sequential single-agent therapy consisting of prednisone followed by vinblastine, cyclophosphamide, and finally procarbazine, with a resulting remission. Studies in 1981 revealed a platelet count of 15,000 per cubic millimeter and abdominal lymphadenopathy due to recurrent Hodgkin's disease. Bone marrow examination showed increased numbers of megakaryocytes, but no involvement with Hodgkin's disease. Serum autoantibodies against glycoprotein IIb/IIIa were demonstrable at a titer of 1:160. The patient was treated with six cycles of cyclophosphamide, vincristine, prednisone, and procarbazine (CMOPP), which led to a complete remission of both Hodgkin's disease and immune thrombocytopenia. Ten years later, she remained in complete remission from both disorders and serum autoantibody was no longer demonstrable.
Encouraged by this patient's response, we have since treated eight patients with refractory chronic idiopathic thrombocytopenic purpura and one patient with severe immune thrombocytopenia and chronic lymphocytic leukemia, using protocols of combination chemotherapy commonly used in lymphoproliferative disorders.
Methods
Ten patients were treated, two patients with immune thrombocytopenia associated with either Hodgkin's disease (Patient 1) or chronic lymphocytic leukemia (Patient 10) and eight patients with refractory chronic idiopathic thrombocytopenic purpura. Patient 7 had autoimmune hemolytic anemia, which had been controlled with splenectomy seven years before the development of chronic idiopathic thrombocytopenic purpura, and Patient 9 had an associated antiphospholipid syndrome with previous fetal loss and periodic deep venous thrombosis requiring warfarin therapy. The patients' histories are summarized in Table 1. There were seven women and three men with an average age of 41.5 years (range, 24 to 69). The duration of the thrombocytopenia ranged from 5 to 110 months, and all patients had platelet counts of less than 5000 per cubic millimeter unless they were receiving some form of therapy. The platelet counts just before the start of combination chemotherapy ranged from 10,000 to 159,000 per cubic millimeter with therapy within the past month consisting of high-dose corticosteroids (Patients 1, 2, and 6), cyclophosphamide (Patients 3, 5, and 8), or intravenous gamma globulin (Patients 4 and 9). Patients 7 and 10 had extremely low platelet counts despite treatment with high-dose corticosteroids and intravenous gamma globulin. All the patients had had petechiae and purpura, and 8 of the 10 had required the transfusion of red cells or platelets. Eight patients had epistaxis and evidence of other severe mucosal bleeding: postoperative bleeding (from cataract surgery) in Patient 2; hematuria in Patient 3; hematuria, gastrointestinal bleeding, and two episodes of central nervous system bleeding in Patient 4; hematuria, gastrointestinal bleeding, and three episodes of central nervous system bleeding in Patient 6; gastrointestinal bleeding and one episode of central nervous system bleeding in Patient 7; gastrointestinal bleeding and postsplenectomy bleeding in Patient 8; deep venous thrombosis due to the antiphospholipid syndrome in Patient 9, when her thrombocytopenia prevented the use of warfarin; and uncontrolled epistaxis requiring transfusion in Patient 10.
|
The reasons for choosing combination chemotherapy in these patients varied. In Patient 1, CMOPP was required to treat her Hodgkin's disease. Corticosteroids and splenectomy had failed in Patient 2, and in view of her age, we thought treatment with cyclophosphamide, vincristine, and prednisone (CVP) would allow us to use lower doses of cyclophosphamide. This patient was treated before the potential benefits of danazol, colchicine, and intravenous gamma globulin were appreciated. Patients 3, 5, and 8 responded temporarily to cyclophosphamide but relapsed when it was stopped. It was thought that this agent carried a substantial risk if it was used continuously over a long period. Patients 4, 9, and 10 responded just to intravenous gamma globulin, but only for a short time, and Patients 2 and 7 did not respond to any of the treatments given. The clinical situation of Patient 10 was critical, and we thought combination chemotherapy was most likely to be helpful.
Before combination chemotherapy began, base-line studies consisted of a complete blood count, blood chemistry tests, and tests for antinuclear antibody and antiplatelet antibody. These studies were repeated periodically during and after therapy.
Chemotherapy consisted of cyclophosphamide (400 to 650 mg per square meter of body-surface area intravenously) on days 1 and 8 and prednisone (40 mg per square meter orally) on days 1 to 14, combined with vincristine (2 mg intravenously) on days 1 and 8 (CVP; Patient 2), vincristine (2 mg) on days 1 and 8 plus procarbazine (100 mg per square meter orally) on days 1 to 14 (CMOPP; Patients 1, 3, and 4), or etoposide (100 mg per square meter intravenously) on days 14, 15, and 16 (CEP; Patients 6 through 10). One patient (Patient 5) received both CMOPP and CEP. The chemotherapy cycles were repeated every 28 days, with occasional delays allowed in the event of severe thrombocytopenia (Patient 7), constitutional symptoms (Patients 8 and 9), and depression (Patient 6). There were no dose adjustments except a decrease in the dose of etoposide to 50 mg per square meter in Patient 8 because of severe constitutional symptoms. On occasion, additional corticosteroids, intravenous gamma globulin, or platelet transfusions were given to control mucosal bleeding.
Responses were defined as complete if the platelet count exceeded 180,000 per cubic millimeter after treatment, as partial if the platelet count exceeded 50,000 per cubic millimeter, and as absent if the platelet count was 50,000 per cubic millimeter or less.
Results
The results of combination chemotherapy are shown in Table 2. Six of the 10 patients (Patients 1, 2, 3, 5, 9, and 10) had complete responses. In four, the responses persisted for periods longer than 11, 30, 54, or 126 months. The fifth patient (Patient 2) responded initially to three cycles of CVP and remained in remission for nine months. After a relapse, she responded to two cycles of CVP and remained in complete remission for an additional 48 months, when she died of a nonhemorrhagic cerebrovascular accident. She had a normal platelet count at the time of her death. The final patient with a complete response (Patient 9) had normal platelet counts after two courses of CEP (maximal count, 356,000 per cubic millimeter), which persisted until just before her fifth course of therapy, when her thrombocytopenia recurred.
|
|
Side effects of therapy were minimal. Mild nausea and vomiting, alopecia, and acne were noted in some patients. Severe malaise required the postponement of therapy on two occasions in Patient 8; the dose of etoposide was decreased to 50 mg per square meter because of these symptoms. Leukopenia was surprisingly uncommon, occurring in just two patients and only during the first course of therapy (1900 and 1500 leukocytes per cubic millimeter in Patients 8 and 9, respectively). No infections were seen.
Discussion
Approximately 80 percent of patients with chronic idiopathic thrombocytopenic purpura respond to prednisone or splenectomy. Traditional second- and third-line agents are useful in some patients with refractory disease, but in many the disease fails to respond to any therapy, causes symptoms, and is associated with a mortality rate of 15 to 20 percent3,6,7,8,9. Therefore, among these patients, a more aggressive approach is warranted.
The complete disappearance of autoantibody-induced immune thrombocytopenia in a patient with recurrent Hodgkin's disease treated with CMOPP suggested to us that combination chemotherapy might be useful in the treatment of refractory chronic idiopathic thrombocytopenic purpura. Anecdotal cases have been previously reported in which this approach has been used11,12.
The results reported here, showing complete responses requiring no further therapy in 5 of 10 patients and a stable partial response in another, provide evidence that combination chemotherapy is beneficial in some patients with chronic idiopathic thrombocytopenic purpura. Corticosteroids and splenectomy had failed in all patients, and three of the six who responded had received multiple agents, including full-dose cyclophosphamide. Complete responses were associated with the disappearance of autoantibodies in three patients and with a stable decrease to about 5 percent of the values obtained at the time of diagnosis in the other two.
Among the remaining four patients, Patient 6 received only two courses of therapy and had a partial response when he decided to forgo further therapy. Patients 4 and 7 had no response. These three patients had particularly severe disease, as shown by refractoriness to multiple previous therapies, severe mucosal bleeding, and history of central nervous system bleeding before combination chemotherapy. All died of their disease. The observation that two of the three had temporary increases in the platelet count (to 61,000 per cubic millimeter in Patient 6 and 46,000 per cubic millimeter in Patient 7) suggests, however, that more aggressive therapy may have had a beneficial effect.
The potential short- and long-term risks of combination chemotherapy must be weighed against the benefits. Early treatment-related mortality in patients with lymphoma is quite low (<5 percent) with similar combinations, such as CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone), CMOPP, and CEPP (cyclophosphamide, etoposide, procarbazine, and prednisone), and most mortality is due to infection associated with neutropenia13,14,15. Neutropenia was uncommon in these patients with idiopathic thrombocytopenic purpura, occurring in the first course of therapy in two patients. There were no infections. Other side effects included mild-to-moderate nausea, alopecia, acne, and malaise. One patient (Patient 6) stopped therapy because of longstanding depression.
Chemotherapy-induced second cancers, particularly acute leukemia, must also be considered16. Acute nonlymphocytic leukemia has been associated with cyclophosphamide,17 procarbazine,18 and etoposide19. The incidence of this disease varies, being influenced by the primary disease and the manner of drug administration. The cumulative incidence of acute leukemia in Hodgkin's disease among patients treated with chemotherapy ranges from 2 percent to 10 percent 10 to 15 years after therapy, whereas patients with breast cancer who receive cyclophosphamide as part of adjuvant therapy have no demonstrable increase in acute leukemia at 10 years16. Acute nonlymphocytic leukemia associated with etoposide therapy occurs frequently in children treated for acute lymphoblastic leukemia (in 12.3 percent at six years), especially in those treated weekly or biweekly, whereas patients receiving the drug twice a month or less have a low frequency of this disease (1.6 percent)19. Secondary lymphomas and solid tumors are regarded as a late complication of therapy in Hodgkin's disease; their occurrence with equal frequency in all treatment schemes indicates that either the disease itself or the radiation is the true cause. There was no increased incidence of secondary lymphomas or solid tumors after adjuvant therapy for breast cancer16.
The risk of secondary cancers in patients with chronic idiopathic thrombocytopenic purpura who receive combination chemotherapy is unknown, but in view of the high mortality rate in patients with refractory idiopathic thrombocytopenic purpura, the risk would seem warranted. Obviously, patients who maintain "safe" platelet counts and have only trivial mucosal bleeding when they are not receiving therapy and patients in whom long-term treatment (e.g., 5 to 10 mg of prednisone per day) is associated with minimal side effects should not be considered as candidates. Picozzi et al.20 reported that many patients with refractory idiopathic thrombocytopenic purpura eventually recover spontaneously. In patients such as those described here, it may not be possible to wait for spontaneous improvement.
Combination chemotherapy appears to benefit some patients with refractory chronic idiopathic thrombocytopenic purpura. Further studies will be required to determine the optimal combination of drugs and timing of therapy and the possibility that more aggressive treatment will be required in some patients with particularly severe disease.
Supported by grants (RR00833 and HL37945) from the U.S. Public Health Service.
Source Information
From the Department of Molecular and Experimental Medicine, the Scripps Research Institute, 10666 N. Torrey Pines Rd., La Jolla, CA 92037, where reprint requests should be addressed to Dr. McMillan.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Immunization of Adults
Rothstein E. P., Bernstein H. H., Long S. S., Gluckman S. J., Krause D. S., Gardner P., Schaffner W.
Extract |
Full Text
N Engl J Med 1993;
329:1046-1048, Sep 30, 1993.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
