- Email: [email protected]
Efficacy and safety of chlorambucil in intractable noninfectious uveitis
Efficacy and Safety of Chlorambucil in Intractable Noninfectious Uveitis The Massachusetts Eye and Ear Infirmary Experience Elisabetta Miserocchi, MD, Stefanos Baltatzis, MD,1 Anthony Ekong, MD, Manolette Roque, MD, C. Stephen Foster, MD, FACS Purpose: To report our experience with the use of chlorambucil for otherwise treatment-resistant uveitis and to assess its safety and efficacy. Design: Noncomparative interventional case series. Participants: Twenty-eight patients with intractable noninfectious uveitis. Methods: We reviewed the records of 28 patients (56 eyes) with chronic noninfectious uveitis who were treated with chlorambucil from 1987 to 2000. Diagnoses included Adamantiades-Behc¸et’s disease (ABD) (7 patients), juvenile rheumatoid arthritis (JRA)–associated uveitis (10 patients), pars planitis (2 patients), sympathetic ophthalmia (1 patient), idiopathic uveitis (6 patients), Crohn’s disease (1 patient), and HLA-B27–associated uveitis (1 patient). All patients were refractory to other immunomodulatory therapy and systemic steroids. The median duration of treatment with chlorambucil was 12 months (range, 4 –50 months), whereas the median daily dosage was 8 mg (range, 4 –22 mg). Patients were followed for a median follow-up period of 46 months (range, 4 –166 months) after chlorambucil treatment was begun and continued to be followed for relapse after cessation of therapy. Main Outcome Measures: Visual outcome, response to treatment, treatment-related side effects, drug dosage, previous and final treatment, discontinuation of systemic corticosteroids. Results: Chlorambucil was discontinued in seven patients because of side effects: two females had temporary amenorrhea develop, two patients had unacceptable gastrointestinal intolerance, one patient had infection, and 2 patients had progressive leukopenia. Nineteen patients (68%) showed positive clinical response to the treatment, four (14%) initially responded then relapsed after discontinuation of the drug, three patients with ABD had improvement of ocular disease but worsening of systemic symptoms, and two had persistent inflammation. Visual acuity was improved in 24 eyes (43%), stable in 22 (39%), and worsened in 10 eyes (18%). Systemic prednisone was successfully discontinued in 19 of the 28 patients (68%), and 14 patients were free of inflammation at the end of follow-up without any systemic medication. Conclusions: Chlorambucil can be a safe and effective alternative for preserving vision in patients with otherwise treatment resistant uveitis. Ophthalmology 2002;109:137–142 © 2002 by the American Academy of Ophthalmology. Chlorambucil is an alkylating immunosuppressive agent, similar to cyclophosphamide, introduced in 1953 as a more stable and less toxic derivative of nitrogen mustard. It is used for the treatment of hematologic, malignant, and rheumatoid diseases, including low-grade non-Hodgkin’s lymphoma, chronic lymphocytic leukemia, Hodgkin’s disease, and amyloidosis complicating juvenile chronic arthritis.1–3 The first reported ophthalmic use of chlorambucil was that of Mamo and Azzam,4 who, in 1970, obtained encourOriginally received: November 6, 2000. Accepted: July 3, 2001. Manuscript no. 200758. From the Immunology and Uveitis Service, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts. Reprint requests to C. Stephen Foster, MD, FACS, Ocular Immunology and Uveitis Service, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, 243 Charles St., Boston, MA 02114. © 2002 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
aging preliminary results in 11 patients with ocular or neural Adamantiades-Behc¸et’s disease (ABD), a potentially blinding and lethal disease. In a follow-up report in 1976,5 long-term remission varying between 23 and 66 months was reported after treatment for approximately 2 years. Subsequently, other investigators 6 – 8 confirmed the efficacy of chlorambucil in the treatment of ABD and attested to its ability to induce long-term remissions.9,10 Chlorambucil has also been used successfully in the treatment of various other forms of uveitis recalcitrant to conventional therapy.11–14 As with most cytotoxic agents, chlorambucil has potential toxic side effects, including bone marrow suppression, gonadal dysfunction, and increased risk of subsequent development of malignancy. Some have, thus, questioned its use because of concerns about these side effects,15–17 whereas others18 have advocated its use. We agree that these side effects need to be mitigated, and we have used techniques to protect fertility in women of ISSN 0161-6420/02/$–see front matter PII S0161-6420(01)00864-8
137
Ophthalmology Volume 109, Number 1, January 2002 childbearing age19,20 during the therapy and have offered sperm banking for postpubescent males. The purpose of this report is to share our experience with the use of chlorambucil for otherwise treatment-resistant uveitis.
Patients and Methods We reviewed the charts of 28 patients (56 eyes) with chronic noninfectious uveitis treated with chlorambucil on the Ocular Immunology and Uveitis Service of the Massachusetts Eye and Ear Infirmary from April 1987 to August 2000. All patients were refractory to other immunomodulatory therapy and systemic steroids. Most patients were referred after extensive treatment elsewhere. Chlorambucil was given to patients after they were informed of its potential side effects, including gonadal dysfunction (azoospermia and amenorrhea). We have always strongly advised adolescent men and young adults who are still interested in procreation to bank their sperm before embarking on cytotoxic therapy. The menstrual and pregnancy histories and oral contraceptive use were noted for women. Ovarian suppression with synthetic gonadotropin-releasing hormone (GnRH-a), such as leuprolide acetate (Lupron) available in subcutaneous or depot formulation (1 and 3 months), was recommended to women of childbearing age. Other available agonists include nafarelin acetate (Synarel), an intranasal spray, and goserelin (Zoladex), a sustained-release formulation (lasting 1–3 months). The use of these synthetic GnRH agonists leads to amenorrhea, with relative estrogen deficiency and sustained decreased folliclestimulating hormone (FSH) and luteinizing hormone (LH) production within 3 to 8 weeks, resulting in “temporary medical castration.” Menses usually return within 2 months of discontinuation of the synthetic GnRH agonist therapy.20 At each visit, women of childbearing age were questioned about irregularity or interruptions in their menstrual cycle and the presence of symptoms of recent ovarian failure, including hot flashes, night sweats, insomnia, vaginal dryness, decreased libido, and bladder infection. When amenorrhea occurred in patients not on concurrent ovarian suppression therapy, chlorambucil was discontinued. Because chlorambucil is a potential teratogen and has been linked to urogenital abnormalities in rats, contraception during chlorambucil therapy was advised. Patients were started on 0.1 mg/kg/day, titrating the dosage on the basis of the clinical response and drug tolerance every 6 weeks. Our main treatment goals were control of all inflammation, including choroidal and retinal inflammation, and macular edema, off all corticosteroid therapy, freedom from relapses during and after therapy with chlorambucil, improvement in visual acuity, and in cases of ABD improvement of systemic symptoms (mainly oral and genital ulcers). Complete hemograms, including platelet levels, leukocyte differentials, and urinalysis were obtained before initiation of therapy, again at first follow-up visit in 2 weeks, and thereafter every 3 to 6 weeks. We aimed for a leukocyte count of not less than 3000 cells/l, a neutrophil count of not less than 1500 cells/l, and a platelet count of above 75,000 cells/l.3 If the blood parameters fell below these levels, the drug was temporarily discontinued or the dosage reduced until the hematologic parameters normalized. Because chlorambucil could not be expected to have reached fully therapeutic levels until 4 to 6 weeks after initiation of treatment, corticosteroid therapy was continued unaltered during the initial weeks of chlorambucil therapy. The steroid dosage was then
138
tapered as patients demonstrated improving clinical response to chlorambucil. After 3 months of therapy, a determination of whether to continue or discontinue therapy was made. Chlorambucil was discontinued or an alternative immunosuppressive agent substituted if an inadequate clinical response was observed during this 3-month time frame. If a good clinical response was obtained, the dosage was reduced progressively in the subsequent 3 to 6 months. Patients continued to be followed for relapse after cessation of therapy.
Results Twenty-six women and two men made up the study group. The average age of the females before chlorambucil therapy was 28 (range, 5– 44); the two males were ages 7 and 70, respectively. Diagnoses included ABD (7 patients), juvenile rheumatoid arthritis–associated uveitis (10 patients), idiopathic uveitis (6 patients), HLA-B27–associated uveitis (1 patient), pars planitis (2 patients), Crohn’s disease (1 patient), and sympathetic ophthalmia (1 patient) (Table 1). The locations of the uveitis were anterior (11 patients), intermediate (2 patients), posterior (4 patients), and panuveitis (11 patients). Table 1 also includes data regarding chlorambucil dosage (initial dose, highest dose, and cumulative dose calculated for each patient). The median duration of treatment with chlorambucil was 12 months (range, 4 –50 months), and the median daily dose was 8 mg (range, 4 –22 mg/day); patients were followed for a median follow-up period of 46 months (range, 4 –166 months) after chlorambucil treatment was begun. Table 2 lists visual outcome, follow-up time from the beginning of chlorambucil treatment, response to treatment, occurrence of treatment-related side effects, and treatment at the end of follow-up. Chlorambucil was discontinued in seven patients (25%); two females who had chosen not to take leuprolide acetate (patients 1 and 2) reported amenorrhea after institution of chlorambucil, two cases had unacceptable gastrointestinal intolerance (patients 8 and 10), patient 22 had sinus infection and shingles; in the other two patients (patients 5 and 21) the drug was discontinued because of progressive leukopenia (Table 2). Patient 1 had her ocular inflammation controlled during the 10 months of treatment with chlorambucil, but the drug was discontinued because of amenorrhea, and she had relapse of inflammation after chlorambucil was discontinued. Patient 2, who was recalcitrant to all available and previous immunosuppressive agents, responded dramatically to 8 mg/day of chlorambucil before amenorrhea developed. The duration of treatment was 6 months. Ocular inflammation relapsed in this patient after chlorambucil was discontinued. Patient 4, who was not on concurrent ovarian suppression therapy, reported temporary amenorrhea (11 months) during follow-up, but subsequently monthly regular menstruations returned. The three females who had amenorrhea develop (patients 1 and 2 while on chlorambucil and patient 4 later on during follow-up) had all been offered ovarian preservation/protection strategies of synthetic GnRH-a therapy, but they had declined this treatment. All these patients had subsequent resumption of regular menstruation. Response to treatment (efficacy), as listed in Table 2, has been divided in: (1) positive response with improvement of ocular inflammation (yes); (2) positive response with improvement of ocular inflammation but subsequent relapse when treatment was discontinued (yes, relapse); (3) improvement of ocular inflammation but worsening of systemic symptoms in patients with ABD (yes eye, no systemic); and (4) lack of response to treatment with continuous active inflammation (no).
Miserocchi et al 䡠 Chlorambucil in Intractable Noninfectious Uveitis Table 1. Patient Characteristics, Diagnosis, Previous Treatment, and Chlorambucil Dosage Patient No. Age/Gender
Diagnosis
Previous Therapy
Initial Dose (mg)
Highest Dose (mg)
Cumulative Dose (mg)
Duration (mos)
1) 42/F 2) 39/F 3) 34/F 4) 36/F 5) 28/F 6) 7/F 7) 29/F 8) 21/F 9) 32/F 10) 7/F 11) 35/F 12) 37/F 13) 15/F 14) 6/M 15) 70/M 16) 17/F 17) 37/F 18) 31/F 19) 28/F 20) 28/F 21) 5/F 22) 37/F 23) 44/F 24) 37/F 25) 44/F 26) 20/F 27) 26/F 28) 22/F
ABD Idiopathic ABD Pars planitis Juvenile rheum arth Juvenile rheum arth Juvenile rheum arth Juvenile rheum arth ABD Juvenile rheum arth Pars planitis Idiopathic Juvenile rheum arth Juvenile rheum arth Sympathetic ophthalmia ABD Idiopathic Juvenile rheum arth Idiopathic Idiopathic Juvenile rheum arth ABD ABD Idiopathic ABD IBD HLA-B27 Juvenile rheum arth
PRD, MTX, AZA, CSA PRD, MTX, CCEPT, CSA, ENB PRD, CTX, CSA PRD, MTX, CSA PRD, MTX, CSA PRD, MTX PRD, MTX PRD, AZA PRD, AZA PRD, MTX PRD, MTX PRD PRD, MTX, AZA PRD, MTX PRD, MTX, CCEPT PRD, AZA, CSA PRD, MTX, CSA PRD, MTX, AZA, CSA PRD, MTX, CSA PRD, CCEPT PRD, MTX PRD, MTX, AZA, CSA PRD, AZA PRD, MTX, CSA, ENB, CCEPT PRD, CTX, CSA, AZA PRD, CSA, AZA PRD, MTX PRN, AZA
12 8 12 6 6 4 4 10 4 4 6 8 8 6 8 6 4 8 6 6 2 6 6 4 14 8 2 4
22 8 16 8 6 6 8 12 14 4 8 12 10 8 12 18 8 10 8 8 4 8 10 8 22 16 4 10
4602 812 6656 2924 900 958 1948 1068 936 544 992 3022 2644 4641 4350 4176 2358 6874 864 1092 690 914 748 1758 4990 1788 294 2160
10 6 50 18 9 6 11 26 16 4 6 10 6 22 13 15 13 22 6 5 5 4 4 7 8 14 4 9
ABD ⫽ Adamantiades-Behc¸et’s disease; AZA ⫽ azathioprine; CCEPT ⫽ mycophenolate mofetil (CellCept); CSA ⫽ cyclosporin; CTX ⫽ cyclophosphamide; ENB ⫽ etanercept (Enbrel); F ⫽ female; IBD ⫽ inflammatory bowel disease; Juvenile rheum arth ⫽ juvenile rhematoid arthritis; M ⫽ male; MTX ⫽ methotrexate; PRD ⫽ prednisone.
Nineteen patients (68%) had a positive clinical response after chlorambucil was instituted; four patients (14%) showed clinical improvement while on treatment but then had relapse of inflammation after the drug was discontinued, necessitating different systemic agents to control ocular disease. Three patients with ABD (patients 9, 22, and 25) had a dramatic ocular response to the treatment but worsening of systemic symptoms (mouth and genital ulcers). Two patients (patients 20 and 27) had persistent inflammation and never responded to chlorambucil treatment. Questionable compliance in the consumption of daily chlorambucil for patient 20 was hypothesized because of the lack of reduction in white blood cell count, which is an inescapable consequence of this treatment. Among the 19 patients who responded to chlorambucil, 12 had stable, quiet eyes at the end of follow-up without any systemic treatment, whereas 3 were placed on other agents because of chlorambucil-induced side effects (patients 2, 5, and 8). Two were still taking chlorambucil (patients 13 and 15), and two required low doses of systemic corticosteroids to maintain ocular disease remission (patients 7 and 18). The average duration of remission of ocular inflammation without relapse was 48 months (range, 9 –150 months). We were able to discontinue the systemic corticosteroids in 19 (68%) of the 28 patients who were initially taking systemic prednisone, although in 9 patients (32%) chronic low doses were still required at the end of follow-up to control ocular inflammation. Twenty-four eyes (43%) showed improvement of visual acuity at the last visit, 22 (39%) were stabilized, whereas 10 eyes (18%) worsened during the follow-up observation period. Our two male patients are not sterile, one female has delivered
a full-term normal baby, and no patient has had a malignancy develop.
Discussion Chlorambucil is an alkylating agent first introduced for treatment of hematologic and malignant diseases1,2 and subsequently, on the basis of its ability to reduce circulating B lymphocytes, found to also be effective in rheumatologic disorders such as amyloidosis complicating juvenile rheumatoid arthritis,3 systemic lupus erythematosus, and nephrotic syndrome.21–23 The relatively scarce use of this agent in immunologic disorders (and in the treatment of vision-threatening conditions caused by intraocular inflammation), has been due to its potential side effects, the most important of which are bone marrow depression, sterility, and secondary malignancies. The documented ophthalmologic importance of chlorambucil is related mostly to treatment of ABD4,5,7,9,10,18,24 and dates from 1970, when it was first reported4 to be effective in controlling ocular inflammation, systemic symptoms, and providing long-term remission for patients with ABD when corticosteroids and more conventional immunosuppressants had failed. There are few reports describing the use of chlorambucil in the treatment of other types of ocular inflammation such as sympathetic ophthalmia,25 juvenile rheumatoid arthritis–
139
Ophthalmology Volume 109, Number 1, January 2002 Table 2. Visual Outcome, Response to Treatment, Treatment-related Side Effects, and Final Therapy Initial Visual Acuity
Final Visual Acuity
Right Eye
Left Eye
Right Eye
Left Eye
Followup (mos.)
1 2 3 4 5 6 7 8
20/200 20/200 20/200 20/80 20/25 20/70 20/60 20/200
20/20 20/200 20/80 20/300 20/200 20/30 20/50 20/400
20/70 20/30 20/70 20/30 20/80 20/40 20/20 20/200
20/40 20/30 20/100 20/300 20/200 20/30 20/20 NLP
18 10 150 24 33 40 26 96
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
20/50 20/20 20/50 20/20 CF HM NLP 20/400 20/25 20/100 20/25 20/30 20/20 20/50 20/60 CF
20/70 20/20 20/70 CF 20/80 20/200 20/60 20/200 20/40 20/50 20/20 20/30 CF 20/30 20/80 20/20
20/25 20/20 20/20 20/20 CF CF NLP 20/200 20/25 20/100 20/40 20/40 20/20 20/50 20/30 CF
20/25 20/20 20/25 CF 20/80 20/25 20/30 20/200 20/30 20/60 20/20 20/40 CF 20/30 20/40 20/20
42 36 39 10 6 48 13 84 54 27 52 5 8 4 84 7
Yes eye, No systemic Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes eye, No systemic Yes, relapse Yes, relapse
25 26 27 28
20/40 NLP 20/40 HM
20/40 20/40 20/60 CF
LP NLP 20/20 20/40
20/20 20/30 20/20 20/30
72 16 114 192
Yes eye, No systemic Yes No Yes
Patient No.
Response* Yes, relapse Yes Yes Yes Yes Yes, relapse Yes Yes
Side Effects *Cystitis, amenorrhea T *Rashes, amenorrhea T — Amenorrhea T *Leukopenia — — *GI upset — *Gl upset — — — — — — — — — — *Leukopenia *Infection — — — — — —
Current Therapy PRD, IFN IV Ig None None CCEPT, ENB CCEPT, CSA PRD MTX, CSA, PRD AZA None None None CHL None CHL, PRD None None PRD None None None PRD None PRD, ENB, CHL PRD, MTX None IV CTX, PRD None
* Response to chlorambucil has been listed as: Yes, patients responded to treatment and did not have relapse of ocular inflammation; Yes, relapse, patient initially had a positive response to treatment but then relapsed after discontinuation of chlorambucil; Yes eye, no systemic; patients with AdamantiadesBehc¸ et’s disease who had improvement of ocular inflammation while taking chlorambucil but did not have improvement of systemic symptoms (i.e., mouth, genital ulcer; no, patients that continued to have active ocular inflammation and never showed clinical response while taking chlorambucil. ABD ⫽ Adamantiades-Behc¸ et’s disease; AZA ⫽ azathiaprine; CF ⫽ counting fingers; CCEPT ⫽ mycophenolate mofetil (CellCept); CHL ⫽ chlorambucil; CSA ⫽ cyclosporin; CTX ⫽ cyclophosphamide; ENB ⫽ Enbrel; HM ⫽ hand motions; IBD ⫽ inflammatory bowel disease; IFN ⫽ interferon; IV ⫽ intravenous; IV Ig ⫽ intravenous immunoglobulin; LP ⫽ light perception; MTX ⫽ methotrexate; NLP ⫽ no light perception; PRD ⫽ prednisone; T ⫽ temporary. Discontinuation of treatment.
associated iridocyclitis,26,27 and intractable idiopathic uveitis.12,18 One of the largest series has been reported by Godfrey et al,11 who described encouraging results in 31 cases of recalcitrant, corticosteroid-resistant uveitis treated with chlorambucil and followed for 10 years; one third of the patients had sustained reduction of ocular inflammation, and relatively few side effects were encountered. Nineteen patients (68%) in our group responded to chlorambucil treatment, and four (14%), although initially responding to the drug, subsequently relapsed when this was discontinued, necessitating resumption of treatment with different agents. Three patients with ABD had improvement of ocular inflammation but worsening of systemic symptoms. Moreover, although most of the patients were corticosteroid resistant, after the chlorambucil was given, we were able to discontinue the systemic corticosteroids in 19 of them. The data regarding the efficacy of chlorambucil in treat-
140
ing ocular ABD are impressive; in a comparison9 of the efficacy of chlorambucil and oral corticosteroids, the patients treated with corticosteroids alone showed recurrence of disease after withdrawal, decreased visual acuity, and one had progression of neurologic symptoms with hemiplegia, whereas patients treated with chlorambucil all had complete remission of disease without further ocular damage or deterioration in visual acuity. The drug-related side effects seen in the ophthalmic literature reports of patients treated with chlorambucil range from mild, such as skin rash (22%), fatigue (25%), anorexia, gastrointestinal upset (7%–25%), and urinary disturbances (9%), to severe, including leukopenia causing secondary infection (19%), thrombocytopenia, and anemia (12%– 15%).11,12 The occurrence of bone marrow depression is, of course, one of the major potential risks of immunosuppression but is also the most preventable problem through frequent monitoring of hematologic parameters and temporary with-
Miserocchi et al 䡠 Chlorambucil in Intractable Noninfectious Uveitis drawal or reduction of dosage. One of our patients had a sinus infection and shingles develop, both of which resolved after discontinuation of chlorambucil and institution of oral acyclovir. None of our patients had fatal complications. Another concern in the use of chlorambucil arises from its potential effects on gonadal function; nitrogen mustard derivatives are known to cause impairment in gonadal function by affecting the most actively proliferating cells (spermatogonia and oocytes), which are susceptible to even low doses of cytotoxic agents.28,29 Sterility as a consequence of alkylating therapy seems to occur in men more frequently than in women, occurring in up to 90% of postpuberal males.20 The loss of germinal cells in males treated with chlorambucil results in sperm count reduction (oligospermia or azoospermia) and secondary effects on the endocrine function of the testes and the hypothalamic-pituitary gonadal axis (increase in serum folliclestimulating hormone (FSH) and luteinizing hormone (LH)). Sperm count is not immediately affected, and the eventual recovery of sperm production depends on the survival and ability of the spermatogenic cells to differentiate.17 There is no relationship between the level of bone marrow depression and reduction of spermatogenesis, whereas there is a well-known correlation between the cumulative dose and the toxic gonadal effect of chlorambucil. Cumulative doses less than 8.2 mg/kg in children and 6.1 mg/kg in adults have been demonstrated to not affect fertility.17,21,22,29 Tabbara15 described effects on spermatogenesis of chlorambucil in 10 males who received a cumulative dose ranging from 10 to 56 mg/kg; 7 had oligospermia and 3 azoospermia, and in 1 there was evidence of biopsy-proven testicular atrophy. The author concluded that the side effects of chlorambucil therapy in patients with ABD outweigh its benefits. Others have vigorously disagreed with this conclusion,18 reasoning that even if spermatogenesis is impaired, it is a small price to pay for prevention of blindness, especially because sperm cryopreservation is so widely and readily available. Both the male patients treated with chlorambucil in our series previously failed to respond to other chemotherapy; one was a 70-year-old man with sympathetic ophthalmia, and the other one was a 6-year-old boy with juvenile rheumatoid arthritis who received the treatment before puberal maturation began. Gonadal toxicity in females is essentially due to destruction of oocytes, resulting in loss of follicles and estrogen production. Cytotoxic therapy can often induce temporary amenorrhea (from a few months up to 3 years), which seems to be independent of age; in contrast, permanent amenorrhea associated with ovarian failure dramatically increases with age.17,20,21 Chlorambucil was discontinued in two of our patients because of amenorrhea, and another one experienced irregular periods later on during the follow-up but subsequently had reinstitution of regular periods. All those patients had been offered our ovarian preservation/protection strategies of synthetic GnRH-a therapy, but they had declined this treatment because of the menopausal side effects to be expected, the increased risk of osteoporosis, and the additional costs associated with the medication and its monitoring. In a study of 26 children treated with chlorambucil for
nephrotic syndrome,21 all 10 females have had normal pubertal development and regular menstruation, and other reports of women treated for malignancies suggest that they have had normal children and normal levels of estradiol, FSH, and LH. One of our patients had a normal pregnancy 4 years after the discontinuation of chlorambucil. After prolonged exposure to alkylating agents, human ovaries show fibrosis and follicle destruction with increased levels of FSH and LH and decreased estradiol. The result of these changes is hypergonadotropic hypogonadism and subsequent amenorrhea, with likely irreversible ovarian dysfunction and infertility. These hormonal and ovarian side effects are more common in older patients. Younger patients tend to tolerate a higher dose of alkylating agents; persistent amenorrhea depends on the age of the patient, cumulative dose, and duration of therapy, with age greater than 30 years and large cumulative dose (⬎300 mg/kg) being worse prognostic indicators. The published data do not permit a clear picture of whether there is a threshold cumulative dose associated with amenorrhea. It is not clear whether the length of amenorrhea after chlorambucil treatment predicts recovery of menses or fertility, nor do we know whether recovery of menses protects a given patient from future early menopause. In view of all the potential effects on gonadal function, it is important to take an accurate endocrinologic history of the patient before starting treatment with chlorambucil and to discuss the possibility of different techniques to protect fertility, such as sperm banking, cryopreservation of germ cells, hormonal treatment, and assisted reproductive technologies.19,20 Treatment with GnRH analogues induces a hypoestrogenic state, resulting in a small loss in bone density in some women over the course of treatment. For a period up to 6 months, this bone loss should not be important, but in patients with major risk factors for decreased bone mineral content this treatment may pose an additional risk. Another important consequence of estrogen deficiency in women with premature menopause is the increased risk for cardiovascular disease. In cases of severe uveitis recalcitrant to other immunosuppressive agents, when progression of the disease will inexorably lead to bilateral blindness, the young adult patient may have to face the difficult decision between blindness and sterility. A possible link between immunosuppressive drugs and malignancies in the rheumatic diseases has always been hypothesized and has generated several controversial opinions. The reports of malignancy developing in patients treated with immunosuppressants suggest an association but not a causation. Other serious side effects of chlorambucil are rare and usually occur after several years of therapy with multipledose regimens; these side effects include myelofibrosis, bone marrow failure, and pulmonary fibrosis. The literature suggests that a number of rheumatologic disorders are inherently associated with an increased risk of malignancy and that those patients have an increased risk of neoplasm of the immune system compared with the general population;30,31 few studies have been able to examine the
141
Ophthalmology Volume 109, Number 1, January 2002 additional risk conferred by immunosuppressant treatment and whether the risk of malignancies is related to the duration of the chemotherapy. Results from the largest European study,32 aimed at determining the relationship between cumulative exposure, duration, and the risk of immune system malignancy, showed that immunosuppressive treatment confers an increased risk of certain types of cancer in patients with rheumatic diseases over and above the risk conferred by the disease itself and that this risk increases with duration of treatment. The increased malignancy risk evaluated in this study was confined to neoplasms of the immune system (lymphomas and leukemias), skin, and bladder. Studies on measurement of chromosomal damage33 have shown that those abnormalities are higher in ABD patients that received cytotoxic agents compared with normal subjects. None of our patients had neoplasia develop, whereas two patients in the series of Godfrey et al11 were diagnosed with sarcoma; but one of these tumors developed before the patient started chlorambucil therapy. In conclusion, with this study we demonstrate that chlorambucil can be considered an effective drug for controlling ocular inflammation in patients with uveitis that is refractory to other less potentially toxic treatment. The power of this drug in dramatically changing the prognosis of potentially blinding diseases such as ABD and also life-threatening conditions like amyloidosis complicating juvenile rheumatoid arthritis has been unquestionably demonstrated. Although chlorambucil is effective in most cases of inflammation, we would emphasize that because of the potential for significant drug-related side effects, it should not be used as a first-line agent, especially in young patients with non–life-threatening diseases. Furthermore, patients treated with this drug should be very carefully selected and monitored by an individual who is, by virtue of training and experience, truly an expert in immunomodulation with this specific drug.
References 1. Portlock CS. Management of the indolent non-Hodgkin’s lymphomas. Semin Oncol 1980;7:292–301. 2. Selby P, Patel P, Milan S, et al. ChlVPP combination therapy for Hodgkin’s disease: long-term results. Br J Cancer 1990; 62:279 – 85. 3. Savolainen HA. Chlorambucil in severe juvenile chronic arthritis: longterm followup with special reference to amyloidosis. J Rheumatol 1999;26:898 –903. 4. Mamo JG, Azzam SA. Treatment of Behc¸ et’s disease with chlorambucil. Arch Ophthalmol 1970;84:446 –50. 5. Mamo JG. Treatment of Behc¸ et disease with chlorambucil: a follow up report. Arch Ophthamol 1976;94:580 –3. 6. Ben Ezra D, Cohen E. Treatment and visual prognosis in Behc¸ et’s disease. Br J Ophthalmol 1986;70:589 –92. 7. Tricoulis D. Treatment of Behc¸ et’s disease with chlorambucil. Br J Ophthalmol 1976;60:55–7. 8. Pivetti Pezzi P, Gasparri V, De Liso P, Catarinelli G. Prognosis in Behc¸ et’s disease. Ann Ophthalmol 1985;17:20 –5. 9. Abdalla MI, El-Din Bahgat N. Long–lasting remission of Behc¸ et’s disease after chlorambucil therapy. Br J Ophthalmol 1973;57:706 –11.
142
10. Elliot TJH, Ballinger WH. Behc¸ et’s syndrome: treatment with chlorambucil. Trans Am Ophthalmol Soc 1984;82:264 – 81. 11. Godfrey WA, Epstein WV, O’Connor GR, et al. The use of chlorambucil in intractable idiopathic uveitis. Am J Ophthalmol 1974;78:415–28. 12. Andrasch RH, Pirofsky B, Burns RP. Immunosuppressive therapy for severe chronic uveitis. Arch Ophthalmol 1978;96: 247–51. 13. Jennings T, Tessler HH. Twenty cases of sympathetic ophthalmia. Br J Ophthalmol 1989;73:140 –5. 14. Kanski JJ. Anterior Uveitis in juvenile rheumatoid arthritis. Arch Ophthalmol 1977;95:1794 –7. 15. Tabbara KF. Chlorambucil in Behc¸ et’s disease. A reappraisal. Ophthalmology 1983;90:906 – 8. 16. Berk PD, Goldberg JD, Silverstein MN, et al. Increased incidence of acute leukemia in polycythemia vera associated with chlorambucil therapy. N Engl J Med 1981;304:441–7. 17. Guesry P, Lenoir G, Broyer M. Gonadal effects of chlorambucil given to prepubertal and pubertal boys for nephrotic syndrome. J Pediatr 1978;92:299 –303. 18. Tessler HH, Jennings T. High-dose short-term chlorambucil for intractable sympathetic ophthalmia and Behc¸ et’s disease. Br J Ophthalmol 1990;74:353–7. 19. Slater CA, Liang MH, McCune JW, et al. Preserving ovarian function in patients receiving cyclophosphamide. Lupus 1999; 8:3–10. 20. Blumenfeld Z, Avivi I, Linn S, et al. Prevention of irreversible chemotherapy-induced ovarian damage in young women with lymphoma by a gonadotrophin-releasing hormone agonist in parallel to chemotherapy. Hum Reprod 1996;11:1620 – 6. 21. Callis L, Nieto J, Vila A, Rende J. Chlorambucil treatment in minimal lesion nephrotic syndrome: a reappraisal of its gonadal toxicity. J Pediatr 1980;97:653– 6. 22. Grupe WE. Chlorambucil in steroid-dependent nephrotic syndrome. J Pediatr 1973;82:598 – 606. 23. Baluarte HJ, Hiner L, Gruskin AB. Chlorambucil dosage in frequently relapsing nephrotic syndrome: a controlled clinical trial. J Pediatr 1978;92:295– 8. 24. Smulders FM, Oosterhuis JA. Treatment of Behcet’s disease with chlorambucil. Ophthalmologica 1975;171:347–52. 25. Yang CS, Liu JH. Chlorambucil therapy in sympathetic ophthalmia. Am J Ophthalmol 1995;119:482– 8. 26. Mehra R, Moore TL, Catalano JD, et al. Chlorambucil in the treatment of iridocyclitis in juvenile rheumatoid arthritis. J Rheumatol 1981;8:141– 4. 27. Palmer RG, Kanski JJ, Ansell BM. Chlorambucil in the treatment of intractable uveitis associated with juvenile chronic arthritis. J Rheumatol 1985;12:967–70. 28. Kumar R, Biggart JD, McEvoy J, McGeown MG. Cyclophosphamide and reproductive function. Lancet 1972;1:1212– 4. 29. Richter P, Calamera JC, Morgenfeld MC, et al. Effect of chlorambucil on spermatogenesis in the human with malignant lymphoma. Cancer 1970;25:1026 –30. 30. Patapanian H, Graham S, Sambrook PN, et al. The oncogenicity of chlorambucil in rheumatoid arthritis. Br J Rheumatol 1988;27:44 –7. 31. Symmons DP. Neoplasia in rheumatoid arthritis [editorial]. J Rheumatol 1988;15:1319 –22. 32. Asten P, Barrett J, Symmons D. Risk of developing certain malignancies is related to duration of immunosuppressive drug exposure in patients with rheumatic diseases. J Rheumatol 1999;26:1705–14. 33. Palmer RG, Dore´ CJ, Denman AM. Chlorambucil-induced chromosome damage to human lymphocytes is dose-dependent and cumulative. Lancet 1984;1:246 –9.
aging preliminary results in 11 patients with ocular or neural Adamantiades-Behc¸et’s disease (ABD), a potentially blinding and lethal disease. In a follow-up report in 1976,5 long-term remission varying between 23 and 66 months was reported after treatment for approximately 2 years. Subsequently, other investigators 6 – 8 confirmed the efficacy of chlorambucil in the treatment of ABD and attested to its ability to induce long-term remissions.9,10 Chlorambucil has also been used successfully in the treatment of various other forms of uveitis recalcitrant to conventional therapy.11–14 As with most cytotoxic agents, chlorambucil has potential toxic side effects, including bone marrow suppression, gonadal dysfunction, and increased risk of subsequent development of malignancy. Some have, thus, questioned its use because of concerns about these side effects,15–17 whereas others18 have advocated its use. We agree that these side effects need to be mitigated, and we have used techniques to protect fertility in women of ISSN 0161-6420/02/$–see front matter PII S0161-6420(01)00864-8
137
Ophthalmology Volume 109, Number 1, January 2002 childbearing age19,20 during the therapy and have offered sperm banking for postpubescent males. The purpose of this report is to share our experience with the use of chlorambucil for otherwise treatment-resistant uveitis.
Patients and Methods We reviewed the charts of 28 patients (56 eyes) with chronic noninfectious uveitis treated with chlorambucil on the Ocular Immunology and Uveitis Service of the Massachusetts Eye and Ear Infirmary from April 1987 to August 2000. All patients were refractory to other immunomodulatory therapy and systemic steroids. Most patients were referred after extensive treatment elsewhere. Chlorambucil was given to patients after they were informed of its potential side effects, including gonadal dysfunction (azoospermia and amenorrhea). We have always strongly advised adolescent men and young adults who are still interested in procreation to bank their sperm before embarking on cytotoxic therapy. The menstrual and pregnancy histories and oral contraceptive use were noted for women. Ovarian suppression with synthetic gonadotropin-releasing hormone (GnRH-a), such as leuprolide acetate (Lupron) available in subcutaneous or depot formulation (1 and 3 months), was recommended to women of childbearing age. Other available agonists include nafarelin acetate (Synarel), an intranasal spray, and goserelin (Zoladex), a sustained-release formulation (lasting 1–3 months). The use of these synthetic GnRH agonists leads to amenorrhea, with relative estrogen deficiency and sustained decreased folliclestimulating hormone (FSH) and luteinizing hormone (LH) production within 3 to 8 weeks, resulting in “temporary medical castration.” Menses usually return within 2 months of discontinuation of the synthetic GnRH agonist therapy.20 At each visit, women of childbearing age were questioned about irregularity or interruptions in their menstrual cycle and the presence of symptoms of recent ovarian failure, including hot flashes, night sweats, insomnia, vaginal dryness, decreased libido, and bladder infection. When amenorrhea occurred in patients not on concurrent ovarian suppression therapy, chlorambucil was discontinued. Because chlorambucil is a potential teratogen and has been linked to urogenital abnormalities in rats, contraception during chlorambucil therapy was advised. Patients were started on 0.1 mg/kg/day, titrating the dosage on the basis of the clinical response and drug tolerance every 6 weeks. Our main treatment goals were control of all inflammation, including choroidal and retinal inflammation, and macular edema, off all corticosteroid therapy, freedom from relapses during and after therapy with chlorambucil, improvement in visual acuity, and in cases of ABD improvement of systemic symptoms (mainly oral and genital ulcers). Complete hemograms, including platelet levels, leukocyte differentials, and urinalysis were obtained before initiation of therapy, again at first follow-up visit in 2 weeks, and thereafter every 3 to 6 weeks. We aimed for a leukocyte count of not less than 3000 cells/l, a neutrophil count of not less than 1500 cells/l, and a platelet count of above 75,000 cells/l.3 If the blood parameters fell below these levels, the drug was temporarily discontinued or the dosage reduced until the hematologic parameters normalized. Because chlorambucil could not be expected to have reached fully therapeutic levels until 4 to 6 weeks after initiation of treatment, corticosteroid therapy was continued unaltered during the initial weeks of chlorambucil therapy. The steroid dosage was then
138
tapered as patients demonstrated improving clinical response to chlorambucil. After 3 months of therapy, a determination of whether to continue or discontinue therapy was made. Chlorambucil was discontinued or an alternative immunosuppressive agent substituted if an inadequate clinical response was observed during this 3-month time frame. If a good clinical response was obtained, the dosage was reduced progressively in the subsequent 3 to 6 months. Patients continued to be followed for relapse after cessation of therapy.
Results Twenty-six women and two men made up the study group. The average age of the females before chlorambucil therapy was 28 (range, 5– 44); the two males were ages 7 and 70, respectively. Diagnoses included ABD (7 patients), juvenile rheumatoid arthritis–associated uveitis (10 patients), idiopathic uveitis (6 patients), HLA-B27–associated uveitis (1 patient), pars planitis (2 patients), Crohn’s disease (1 patient), and sympathetic ophthalmia (1 patient) (Table 1). The locations of the uveitis were anterior (11 patients), intermediate (2 patients), posterior (4 patients), and panuveitis (11 patients). Table 1 also includes data regarding chlorambucil dosage (initial dose, highest dose, and cumulative dose calculated for each patient). The median duration of treatment with chlorambucil was 12 months (range, 4 –50 months), and the median daily dose was 8 mg (range, 4 –22 mg/day); patients were followed for a median follow-up period of 46 months (range, 4 –166 months) after chlorambucil treatment was begun. Table 2 lists visual outcome, follow-up time from the beginning of chlorambucil treatment, response to treatment, occurrence of treatment-related side effects, and treatment at the end of follow-up. Chlorambucil was discontinued in seven patients (25%); two females who had chosen not to take leuprolide acetate (patients 1 and 2) reported amenorrhea after institution of chlorambucil, two cases had unacceptable gastrointestinal intolerance (patients 8 and 10), patient 22 had sinus infection and shingles; in the other two patients (patients 5 and 21) the drug was discontinued because of progressive leukopenia (Table 2). Patient 1 had her ocular inflammation controlled during the 10 months of treatment with chlorambucil, but the drug was discontinued because of amenorrhea, and she had relapse of inflammation after chlorambucil was discontinued. Patient 2, who was recalcitrant to all available and previous immunosuppressive agents, responded dramatically to 8 mg/day of chlorambucil before amenorrhea developed. The duration of treatment was 6 months. Ocular inflammation relapsed in this patient after chlorambucil was discontinued. Patient 4, who was not on concurrent ovarian suppression therapy, reported temporary amenorrhea (11 months) during follow-up, but subsequently monthly regular menstruations returned. The three females who had amenorrhea develop (patients 1 and 2 while on chlorambucil and patient 4 later on during follow-up) had all been offered ovarian preservation/protection strategies of synthetic GnRH-a therapy, but they had declined this treatment. All these patients had subsequent resumption of regular menstruation. Response to treatment (efficacy), as listed in Table 2, has been divided in: (1) positive response with improvement of ocular inflammation (yes); (2) positive response with improvement of ocular inflammation but subsequent relapse when treatment was discontinued (yes, relapse); (3) improvement of ocular inflammation but worsening of systemic symptoms in patients with ABD (yes eye, no systemic); and (4) lack of response to treatment with continuous active inflammation (no).
Miserocchi et al 䡠 Chlorambucil in Intractable Noninfectious Uveitis Table 1. Patient Characteristics, Diagnosis, Previous Treatment, and Chlorambucil Dosage Patient No. Age/Gender
Diagnosis
Previous Therapy
Initial Dose (mg)
Highest Dose (mg)
Cumulative Dose (mg)
Duration (mos)
1) 42/F 2) 39/F 3) 34/F 4) 36/F 5) 28/F 6) 7/F 7) 29/F 8) 21/F 9) 32/F 10) 7/F 11) 35/F 12) 37/F 13) 15/F 14) 6/M 15) 70/M 16) 17/F 17) 37/F 18) 31/F 19) 28/F 20) 28/F 21) 5/F 22) 37/F 23) 44/F 24) 37/F 25) 44/F 26) 20/F 27) 26/F 28) 22/F
ABD Idiopathic ABD Pars planitis Juvenile rheum arth Juvenile rheum arth Juvenile rheum arth Juvenile rheum arth ABD Juvenile rheum arth Pars planitis Idiopathic Juvenile rheum arth Juvenile rheum arth Sympathetic ophthalmia ABD Idiopathic Juvenile rheum arth Idiopathic Idiopathic Juvenile rheum arth ABD ABD Idiopathic ABD IBD HLA-B27 Juvenile rheum arth
PRD, MTX, AZA, CSA PRD, MTX, CCEPT, CSA, ENB PRD, CTX, CSA PRD, MTX, CSA PRD, MTX, CSA PRD, MTX PRD, MTX PRD, AZA PRD, AZA PRD, MTX PRD, MTX PRD PRD, MTX, AZA PRD, MTX PRD, MTX, CCEPT PRD, AZA, CSA PRD, MTX, CSA PRD, MTX, AZA, CSA PRD, MTX, CSA PRD, CCEPT PRD, MTX PRD, MTX, AZA, CSA PRD, AZA PRD, MTX, CSA, ENB, CCEPT PRD, CTX, CSA, AZA PRD, CSA, AZA PRD, MTX PRN, AZA
12 8 12 6 6 4 4 10 4 4 6 8 8 6 8 6 4 8 6 6 2 6 6 4 14 8 2 4
22 8 16 8 6 6 8 12 14 4 8 12 10 8 12 18 8 10 8 8 4 8 10 8 22 16 4 10
4602 812 6656 2924 900 958 1948 1068 936 544 992 3022 2644 4641 4350 4176 2358 6874 864 1092 690 914 748 1758 4990 1788 294 2160
10 6 50 18 9 6 11 26 16 4 6 10 6 22 13 15 13 22 6 5 5 4 4 7 8 14 4 9
ABD ⫽ Adamantiades-Behc¸et’s disease; AZA ⫽ azathioprine; CCEPT ⫽ mycophenolate mofetil (CellCept); CSA ⫽ cyclosporin; CTX ⫽ cyclophosphamide; ENB ⫽ etanercept (Enbrel); F ⫽ female; IBD ⫽ inflammatory bowel disease; Juvenile rheum arth ⫽ juvenile rhematoid arthritis; M ⫽ male; MTX ⫽ methotrexate; PRD ⫽ prednisone.
Nineteen patients (68%) had a positive clinical response after chlorambucil was instituted; four patients (14%) showed clinical improvement while on treatment but then had relapse of inflammation after the drug was discontinued, necessitating different systemic agents to control ocular disease. Three patients with ABD (patients 9, 22, and 25) had a dramatic ocular response to the treatment but worsening of systemic symptoms (mouth and genital ulcers). Two patients (patients 20 and 27) had persistent inflammation and never responded to chlorambucil treatment. Questionable compliance in the consumption of daily chlorambucil for patient 20 was hypothesized because of the lack of reduction in white blood cell count, which is an inescapable consequence of this treatment. Among the 19 patients who responded to chlorambucil, 12 had stable, quiet eyes at the end of follow-up without any systemic treatment, whereas 3 were placed on other agents because of chlorambucil-induced side effects (patients 2, 5, and 8). Two were still taking chlorambucil (patients 13 and 15), and two required low doses of systemic corticosteroids to maintain ocular disease remission (patients 7 and 18). The average duration of remission of ocular inflammation without relapse was 48 months (range, 9 –150 months). We were able to discontinue the systemic corticosteroids in 19 (68%) of the 28 patients who were initially taking systemic prednisone, although in 9 patients (32%) chronic low doses were still required at the end of follow-up to control ocular inflammation. Twenty-four eyes (43%) showed improvement of visual acuity at the last visit, 22 (39%) were stabilized, whereas 10 eyes (18%) worsened during the follow-up observation period. Our two male patients are not sterile, one female has delivered
a full-term normal baby, and no patient has had a malignancy develop.
Discussion Chlorambucil is an alkylating agent first introduced for treatment of hematologic and malignant diseases1,2 and subsequently, on the basis of its ability to reduce circulating B lymphocytes, found to also be effective in rheumatologic disorders such as amyloidosis complicating juvenile rheumatoid arthritis,3 systemic lupus erythematosus, and nephrotic syndrome.21–23 The relatively scarce use of this agent in immunologic disorders (and in the treatment of vision-threatening conditions caused by intraocular inflammation), has been due to its potential side effects, the most important of which are bone marrow depression, sterility, and secondary malignancies. The documented ophthalmologic importance of chlorambucil is related mostly to treatment of ABD4,5,7,9,10,18,24 and dates from 1970, when it was first reported4 to be effective in controlling ocular inflammation, systemic symptoms, and providing long-term remission for patients with ABD when corticosteroids and more conventional immunosuppressants had failed. There are few reports describing the use of chlorambucil in the treatment of other types of ocular inflammation such as sympathetic ophthalmia,25 juvenile rheumatoid arthritis–
139
Ophthalmology Volume 109, Number 1, January 2002 Table 2. Visual Outcome, Response to Treatment, Treatment-related Side Effects, and Final Therapy Initial Visual Acuity
Final Visual Acuity
Right Eye
Left Eye
Right Eye
Left Eye
Followup (mos.)
1 2 3 4 5 6 7 8
20/200 20/200 20/200 20/80 20/25 20/70 20/60 20/200
20/20 20/200 20/80 20/300 20/200 20/30 20/50 20/400
20/70 20/30 20/70 20/30 20/80 20/40 20/20 20/200
20/40 20/30 20/100 20/300 20/200 20/30 20/20 NLP
18 10 150 24 33 40 26 96
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
20/50 20/20 20/50 20/20 CF HM NLP 20/400 20/25 20/100 20/25 20/30 20/20 20/50 20/60 CF
20/70 20/20 20/70 CF 20/80 20/200 20/60 20/200 20/40 20/50 20/20 20/30 CF 20/30 20/80 20/20
20/25 20/20 20/20 20/20 CF CF NLP 20/200 20/25 20/100 20/40 20/40 20/20 20/50 20/30 CF
20/25 20/20 20/25 CF 20/80 20/25 20/30 20/200 20/30 20/60 20/20 20/40 CF 20/30 20/40 20/20
42 36 39 10 6 48 13 84 54 27 52 5 8 4 84 7
Yes eye, No systemic Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes eye, No systemic Yes, relapse Yes, relapse
25 26 27 28
20/40 NLP 20/40 HM
20/40 20/40 20/60 CF
LP NLP 20/20 20/40
20/20 20/30 20/20 20/30
72 16 114 192
Yes eye, No systemic Yes No Yes
Patient No.
Response* Yes, relapse Yes Yes Yes Yes Yes, relapse Yes Yes
Side Effects *Cystitis, amenorrhea T *Rashes, amenorrhea T — Amenorrhea T *Leukopenia — — *GI upset — *Gl upset — — — — — — — — — — *Leukopenia *Infection — — — — — —
Current Therapy PRD, IFN IV Ig None None CCEPT, ENB CCEPT, CSA PRD MTX, CSA, PRD AZA None None None CHL None CHL, PRD None None PRD None None None PRD None PRD, ENB, CHL PRD, MTX None IV CTX, PRD None
* Response to chlorambucil has been listed as: Yes, patients responded to treatment and did not have relapse of ocular inflammation; Yes, relapse, patient initially had a positive response to treatment but then relapsed after discontinuation of chlorambucil; Yes eye, no systemic; patients with AdamantiadesBehc¸ et’s disease who had improvement of ocular inflammation while taking chlorambucil but did not have improvement of systemic symptoms (i.e., mouth, genital ulcer; no, patients that continued to have active ocular inflammation and never showed clinical response while taking chlorambucil. ABD ⫽ Adamantiades-Behc¸ et’s disease; AZA ⫽ azathiaprine; CF ⫽ counting fingers; CCEPT ⫽ mycophenolate mofetil (CellCept); CHL ⫽ chlorambucil; CSA ⫽ cyclosporin; CTX ⫽ cyclophosphamide; ENB ⫽ Enbrel; HM ⫽ hand motions; IBD ⫽ inflammatory bowel disease; IFN ⫽ interferon; IV ⫽ intravenous; IV Ig ⫽ intravenous immunoglobulin; LP ⫽ light perception; MTX ⫽ methotrexate; NLP ⫽ no light perception; PRD ⫽ prednisone; T ⫽ temporary. Discontinuation of treatment.
associated iridocyclitis,26,27 and intractable idiopathic uveitis.12,18 One of the largest series has been reported by Godfrey et al,11 who described encouraging results in 31 cases of recalcitrant, corticosteroid-resistant uveitis treated with chlorambucil and followed for 10 years; one third of the patients had sustained reduction of ocular inflammation, and relatively few side effects were encountered. Nineteen patients (68%) in our group responded to chlorambucil treatment, and four (14%), although initially responding to the drug, subsequently relapsed when this was discontinued, necessitating resumption of treatment with different agents. Three patients with ABD had improvement of ocular inflammation but worsening of systemic symptoms. Moreover, although most of the patients were corticosteroid resistant, after the chlorambucil was given, we were able to discontinue the systemic corticosteroids in 19 of them. The data regarding the efficacy of chlorambucil in treat-
140
ing ocular ABD are impressive; in a comparison9 of the efficacy of chlorambucil and oral corticosteroids, the patients treated with corticosteroids alone showed recurrence of disease after withdrawal, decreased visual acuity, and one had progression of neurologic symptoms with hemiplegia, whereas patients treated with chlorambucil all had complete remission of disease without further ocular damage or deterioration in visual acuity. The drug-related side effects seen in the ophthalmic literature reports of patients treated with chlorambucil range from mild, such as skin rash (22%), fatigue (25%), anorexia, gastrointestinal upset (7%–25%), and urinary disturbances (9%), to severe, including leukopenia causing secondary infection (19%), thrombocytopenia, and anemia (12%– 15%).11,12 The occurrence of bone marrow depression is, of course, one of the major potential risks of immunosuppression but is also the most preventable problem through frequent monitoring of hematologic parameters and temporary with-
Miserocchi et al 䡠 Chlorambucil in Intractable Noninfectious Uveitis drawal or reduction of dosage. One of our patients had a sinus infection and shingles develop, both of which resolved after discontinuation of chlorambucil and institution of oral acyclovir. None of our patients had fatal complications. Another concern in the use of chlorambucil arises from its potential effects on gonadal function; nitrogen mustard derivatives are known to cause impairment in gonadal function by affecting the most actively proliferating cells (spermatogonia and oocytes), which are susceptible to even low doses of cytotoxic agents.28,29 Sterility as a consequence of alkylating therapy seems to occur in men more frequently than in women, occurring in up to 90% of postpuberal males.20 The loss of germinal cells in males treated with chlorambucil results in sperm count reduction (oligospermia or azoospermia) and secondary effects on the endocrine function of the testes and the hypothalamic-pituitary gonadal axis (increase in serum folliclestimulating hormone (FSH) and luteinizing hormone (LH)). Sperm count is not immediately affected, and the eventual recovery of sperm production depends on the survival and ability of the spermatogenic cells to differentiate.17 There is no relationship between the level of bone marrow depression and reduction of spermatogenesis, whereas there is a well-known correlation between the cumulative dose and the toxic gonadal effect of chlorambucil. Cumulative doses less than 8.2 mg/kg in children and 6.1 mg/kg in adults have been demonstrated to not affect fertility.17,21,22,29 Tabbara15 described effects on spermatogenesis of chlorambucil in 10 males who received a cumulative dose ranging from 10 to 56 mg/kg; 7 had oligospermia and 3 azoospermia, and in 1 there was evidence of biopsy-proven testicular atrophy. The author concluded that the side effects of chlorambucil therapy in patients with ABD outweigh its benefits. Others have vigorously disagreed with this conclusion,18 reasoning that even if spermatogenesis is impaired, it is a small price to pay for prevention of blindness, especially because sperm cryopreservation is so widely and readily available. Both the male patients treated with chlorambucil in our series previously failed to respond to other chemotherapy; one was a 70-year-old man with sympathetic ophthalmia, and the other one was a 6-year-old boy with juvenile rheumatoid arthritis who received the treatment before puberal maturation began. Gonadal toxicity in females is essentially due to destruction of oocytes, resulting in loss of follicles and estrogen production. Cytotoxic therapy can often induce temporary amenorrhea (from a few months up to 3 years), which seems to be independent of age; in contrast, permanent amenorrhea associated with ovarian failure dramatically increases with age.17,20,21 Chlorambucil was discontinued in two of our patients because of amenorrhea, and another one experienced irregular periods later on during the follow-up but subsequently had reinstitution of regular periods. All those patients had been offered our ovarian preservation/protection strategies of synthetic GnRH-a therapy, but they had declined this treatment because of the menopausal side effects to be expected, the increased risk of osteoporosis, and the additional costs associated with the medication and its monitoring. In a study of 26 children treated with chlorambucil for
nephrotic syndrome,21 all 10 females have had normal pubertal development and regular menstruation, and other reports of women treated for malignancies suggest that they have had normal children and normal levels of estradiol, FSH, and LH. One of our patients had a normal pregnancy 4 years after the discontinuation of chlorambucil. After prolonged exposure to alkylating agents, human ovaries show fibrosis and follicle destruction with increased levels of FSH and LH and decreased estradiol. The result of these changes is hypergonadotropic hypogonadism and subsequent amenorrhea, with likely irreversible ovarian dysfunction and infertility. These hormonal and ovarian side effects are more common in older patients. Younger patients tend to tolerate a higher dose of alkylating agents; persistent amenorrhea depends on the age of the patient, cumulative dose, and duration of therapy, with age greater than 30 years and large cumulative dose (⬎300 mg/kg) being worse prognostic indicators. The published data do not permit a clear picture of whether there is a threshold cumulative dose associated with amenorrhea. It is not clear whether the length of amenorrhea after chlorambucil treatment predicts recovery of menses or fertility, nor do we know whether recovery of menses protects a given patient from future early menopause. In view of all the potential effects on gonadal function, it is important to take an accurate endocrinologic history of the patient before starting treatment with chlorambucil and to discuss the possibility of different techniques to protect fertility, such as sperm banking, cryopreservation of germ cells, hormonal treatment, and assisted reproductive technologies.19,20 Treatment with GnRH analogues induces a hypoestrogenic state, resulting in a small loss in bone density in some women over the course of treatment. For a period up to 6 months, this bone loss should not be important, but in patients with major risk factors for decreased bone mineral content this treatment may pose an additional risk. Another important consequence of estrogen deficiency in women with premature menopause is the increased risk for cardiovascular disease. In cases of severe uveitis recalcitrant to other immunosuppressive agents, when progression of the disease will inexorably lead to bilateral blindness, the young adult patient may have to face the difficult decision between blindness and sterility. A possible link between immunosuppressive drugs and malignancies in the rheumatic diseases has always been hypothesized and has generated several controversial opinions. The reports of malignancy developing in patients treated with immunosuppressants suggest an association but not a causation. Other serious side effects of chlorambucil are rare and usually occur after several years of therapy with multipledose regimens; these side effects include myelofibrosis, bone marrow failure, and pulmonary fibrosis. The literature suggests that a number of rheumatologic disorders are inherently associated with an increased risk of malignancy and that those patients have an increased risk of neoplasm of the immune system compared with the general population;30,31 few studies have been able to examine the
141
Ophthalmology Volume 109, Number 1, January 2002 additional risk conferred by immunosuppressant treatment and whether the risk of malignancies is related to the duration of the chemotherapy. Results from the largest European study,32 aimed at determining the relationship between cumulative exposure, duration, and the risk of immune system malignancy, showed that immunosuppressive treatment confers an increased risk of certain types of cancer in patients with rheumatic diseases over and above the risk conferred by the disease itself and that this risk increases with duration of treatment. The increased malignancy risk evaluated in this study was confined to neoplasms of the immune system (lymphomas and leukemias), skin, and bladder. Studies on measurement of chromosomal damage33 have shown that those abnormalities are higher in ABD patients that received cytotoxic agents compared with normal subjects. None of our patients had neoplasia develop, whereas two patients in the series of Godfrey et al11 were diagnosed with sarcoma; but one of these tumors developed before the patient started chlorambucil therapy. In conclusion, with this study we demonstrate that chlorambucil can be considered an effective drug for controlling ocular inflammation in patients with uveitis that is refractory to other less potentially toxic treatment. The power of this drug in dramatically changing the prognosis of potentially blinding diseases such as ABD and also life-threatening conditions like amyloidosis complicating juvenile rheumatoid arthritis has been unquestionably demonstrated. Although chlorambucil is effective in most cases of inflammation, we would emphasize that because of the potential for significant drug-related side effects, it should not be used as a first-line agent, especially in young patients with non–life-threatening diseases. Furthermore, patients treated with this drug should be very carefully selected and monitored by an individual who is, by virtue of training and experience, truly an expert in immunomodulation with this specific drug.
References 1. Portlock CS. Management of the indolent non-Hodgkin’s lymphomas. Semin Oncol 1980;7:292–301. 2. Selby P, Patel P, Milan S, et al. ChlVPP combination therapy for Hodgkin’s disease: long-term results. Br J Cancer 1990; 62:279 – 85. 3. Savolainen HA. Chlorambucil in severe juvenile chronic arthritis: longterm followup with special reference to amyloidosis. J Rheumatol 1999;26:898 –903. 4. Mamo JG, Azzam SA. Treatment of Behc¸ et’s disease with chlorambucil. Arch Ophthalmol 1970;84:446 –50. 5. Mamo JG. Treatment of Behc¸ et disease with chlorambucil: a follow up report. Arch Ophthamol 1976;94:580 –3. 6. Ben Ezra D, Cohen E. Treatment and visual prognosis in Behc¸ et’s disease. Br J Ophthalmol 1986;70:589 –92. 7. Tricoulis D. Treatment of Behc¸ et’s disease with chlorambucil. Br J Ophthalmol 1976;60:55–7. 8. Pivetti Pezzi P, Gasparri V, De Liso P, Catarinelli G. Prognosis in Behc¸ et’s disease. Ann Ophthalmol 1985;17:20 –5. 9. Abdalla MI, El-Din Bahgat N. Long–lasting remission of Behc¸ et’s disease after chlorambucil therapy. Br J Ophthalmol 1973;57:706 –11.
142
10. Elliot TJH, Ballinger WH. Behc¸ et’s syndrome: treatment with chlorambucil. Trans Am Ophthalmol Soc 1984;82:264 – 81. 11. Godfrey WA, Epstein WV, O’Connor GR, et al. The use of chlorambucil in intractable idiopathic uveitis. Am J Ophthalmol 1974;78:415–28. 12. Andrasch RH, Pirofsky B, Burns RP. Immunosuppressive therapy for severe chronic uveitis. Arch Ophthalmol 1978;96: 247–51. 13. Jennings T, Tessler HH. Twenty cases of sympathetic ophthalmia. Br J Ophthalmol 1989;73:140 –5. 14. Kanski JJ. Anterior Uveitis in juvenile rheumatoid arthritis. Arch Ophthalmol 1977;95:1794 –7. 15. Tabbara KF. Chlorambucil in Behc¸ et’s disease. A reappraisal. Ophthalmology 1983;90:906 – 8. 16. Berk PD, Goldberg JD, Silverstein MN, et al. Increased incidence of acute leukemia in polycythemia vera associated with chlorambucil therapy. N Engl J Med 1981;304:441–7. 17. Guesry P, Lenoir G, Broyer M. Gonadal effects of chlorambucil given to prepubertal and pubertal boys for nephrotic syndrome. J Pediatr 1978;92:299 –303. 18. Tessler HH, Jennings T. High-dose short-term chlorambucil for intractable sympathetic ophthalmia and Behc¸ et’s disease. Br J Ophthalmol 1990;74:353–7. 19. Slater CA, Liang MH, McCune JW, et al. Preserving ovarian function in patients receiving cyclophosphamide. Lupus 1999; 8:3–10. 20. Blumenfeld Z, Avivi I, Linn S, et al. Prevention of irreversible chemotherapy-induced ovarian damage in young women with lymphoma by a gonadotrophin-releasing hormone agonist in parallel to chemotherapy. Hum Reprod 1996;11:1620 – 6. 21. Callis L, Nieto J, Vila A, Rende J. Chlorambucil treatment in minimal lesion nephrotic syndrome: a reappraisal of its gonadal toxicity. J Pediatr 1980;97:653– 6. 22. Grupe WE. Chlorambucil in steroid-dependent nephrotic syndrome. J Pediatr 1973;82:598 – 606. 23. Baluarte HJ, Hiner L, Gruskin AB. Chlorambucil dosage in frequently relapsing nephrotic syndrome: a controlled clinical trial. J Pediatr 1978;92:295– 8. 24. Smulders FM, Oosterhuis JA. Treatment of Behcet’s disease with chlorambucil. Ophthalmologica 1975;171:347–52. 25. Yang CS, Liu JH. Chlorambucil therapy in sympathetic ophthalmia. Am J Ophthalmol 1995;119:482– 8. 26. Mehra R, Moore TL, Catalano JD, et al. Chlorambucil in the treatment of iridocyclitis in juvenile rheumatoid arthritis. J Rheumatol 1981;8:141– 4. 27. Palmer RG, Kanski JJ, Ansell BM. Chlorambucil in the treatment of intractable uveitis associated with juvenile chronic arthritis. J Rheumatol 1985;12:967–70. 28. Kumar R, Biggart JD, McEvoy J, McGeown MG. Cyclophosphamide and reproductive function. Lancet 1972;1:1212– 4. 29. Richter P, Calamera JC, Morgenfeld MC, et al. Effect of chlorambucil on spermatogenesis in the human with malignant lymphoma. Cancer 1970;25:1026 –30. 30. Patapanian H, Graham S, Sambrook PN, et al. The oncogenicity of chlorambucil in rheumatoid arthritis. Br J Rheumatol 1988;27:44 –7. 31. Symmons DP. Neoplasia in rheumatoid arthritis [editorial]. J Rheumatol 1988;15:1319 –22. 32. Asten P, Barrett J, Symmons D. Risk of developing certain malignancies is related to duration of immunosuppressive drug exposure in patients with rheumatic diseases. J Rheumatol 1999;26:1705–14. 33. Palmer RG, Dore´ CJ, Denman AM. Chlorambucil-induced chromosome damage to human lymphocytes is dose-dependent and cumulative. Lancet 1984;1:246 –9.