Cataracts after total body irradiation and bone marrow transplantation in patients with acute leukemia in complete remission: a study of the european group for blood and marrow transplantation

Int. J. Radiation Oncology Biol. Phys., Vol. 41, No. 3, pp. 659 – 668, 1998 Copyright © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/98 $19.00 1 .00

PII S0360-3016(98)00077-7

●

Clinical Investigation CATARACTS AFTER TOTAL BODY IRRADIATION AND BONE MARROW TRANSPLANTATION IN PATIENTS WITH ACUTE LEUKEMIA IN COMPLETE REMISSION: A STUDY OF THE EUROPEAN GROUP FOR BLOOD AND MARROW TRANSPLANTATION YAZID BELKACEMI, M.D.,* MYRIAM LABOPIN, M.D., M.S.,* JEAN-PAUL VERNANT, M.D.,* HANS G. PRENTICE, M.D.,† ANDRE TICHELLI, M.D.,†* ANTON SCHATTENBERG, M.D.,†† MARC A. BOOGAERTS, M.D.,** PETER ERNST, M.D.,§ ALDO DELLA VOLPE, M.D.,‡ ANTONY H. GOLDSTONE, M.D.,*‡ JEAN-PIERRE JOUET, M.D.,§§ LEO F. VERDONCK, M.D.,§† ANNA LOCASCIULLI, M.D.,§‡ BERNARD RIO, M.D.,* MAHMUT OZSAHIN, M.D., PH.D.‡‡ AND NORBERT C. GORIN, M.D.*

Acute Leukemia Working Party of the European Group for Blood and Marrow Transplantation (EBMT). *Paris (France); †Hampstead, London (UK); †*Basel (Switzerland); ††Nijmegen (The Netherlands); **Leuven (Belgium); §Riyadh (Saudi Arabia); ‡Milano (Italy); *‡London (UK); §§Lille (France); §†Utrecht (The Netherlands); §‡Monza (Italy); ‡‡Lausanne (Switzerland) Purpose: Advances in bone marrow transplantation (BMT) have consistently improved long-term survival. Therefore, evaluation of late complications such as cataracts is of paramount importance. Methods and Materials: We analyzed data of 2149 patients from the EBMT registry. A cohort of 1063 patients were evaluable for survival and ophthalmologic status after transplant for acute leukemia (AL) in first or second complete remission. Conditioning therapy included either single-dose total body irradiation (STBI) or fractionated TBI (FTBI) grouped in different dose rates (low: LDR < 0.04 Gy/min; high: HDR > 0.04 Gy/min). Results: The overall 10-year estimated cataract incidence (ECI) was 50%. It was 60% in the STBI group, 43% in the FTBI group < 6 fractions, and 7% in the FTBI group > 6 fractions (p < 1024). It was significantly lower (30%) in the LDR than in the HDR groups (59%; p < 1024). Patients receiving heparin for veno-occlusive disease prophylaxis had fewer cataracts than those who did not (10-year ECI: 33% vs. 53%, respectively; p 5 0.04). The 10-year ECI was 65% in the allogeneic vs. 46% in the autologous BMT patients (p 5 0.0018). Factors independently associated with an increased risk of cataract were an older age (> 23 years), higher dose rate (> 0.04 Gy/min), allogeneic BMT, and steroid administration (> 100 days). The use of FTBI was associated with a decreased risk of cataract. Heparin administration was a protective factor in patients receiving STBI. In terms of cataract surgery, the unfavorable factors for requiring surgery were: age > 23 yr, STBI, dose rate > 0.04 Gy/min, chronic graft-vs.-host disease (cGvHD), and absence of heparin administration. Among the patients who required cataract surgery (111 out of 257), secondary posterior capsular opacification was observed in 15.7%. Conclusion: High dose rate and STBI are the main risk factors for cataract development and the need for surgery, and the administration of heparin has a protective role in cataractogenesis. © 1998 Elsevier Science Inc. Cataract, Posterior capsular opacification, Total body irradiation (TBI), Fractionation, Dose rate, Bone marrow transplantation (BMT), Heparin.

INTRODUCTION Advances in bone marrow transplantation (BMT) over the past 20 years have improved long-term survival of children and adults treated for hematological malignancies. Evaluation of late complications following BMT is of paramount importance. Conditioning regimens including or not total body irradiation (TBI) are involved in late effects following BMT. These include cataract formation, infertility,

growth and cognitive impairment in children, and hypothyroidism (1– 4). Several factors have been reported to influence cataractogenesis after BMT; however, the results in terms of incidence and risk for requiring cataract surgery are controversial because of the heterogeneity of series in terms of ophthalmologic follow-up and survival, and the absence of a well-standardized system for classification and grading of cataract. The objective of this retrospective analysis of the Euro-

Reprint requests to: Yazid Belkace´mi, M.D., Service d’Oncologie-Radiothe´rapie, Hoˆpital Tenon, 4 rue de la Chine,

F-75970 Paris 20, France. Accepted for publication 9 February 1998. 659

660

I. J. Radiation Oncology

●

Biology

●

Physics

pean Group for Blood and Marrow Transplantation (EBMT) is to determine the risk factors for cataract formation, and the need for cataract surgery following TBI and BMT to treat acute leukemia (AL) in first and second complete remission (CR). Moreover, we wanted to investigate the secondary lens opacification after extracapsular surgery (ECS) and intraocular implant because of the possible inhibiting effect of ionizing radiation on secondary lens epithelial cells proliferation after surgery as compared to patients treated for a senile cataract (5, 6). METHODS AND MATERIALS As of June 1996, among the 2149 EBMT questionnaires, 1760 patients were evaluable for survival. The questionnaires included all information concerning TBI parameters, cataract and its surgical treatment as well as data concerning the secondary lens opacification after ECS and intraocular lens implantation. A total of 1094 patients survived more than 1 year, including five patients who developed a cataract during the first year after BMT. In 31 cases, no ophthalmologic follow-up was recorded. Therefore, the analysis concerned an homogeneous population of 1063 patients with AL grafted in CR1 or CR2. Patients’ characteristics are presented in Table 1. The median age of the population was 23 years (mean: 25 6 12, range: 1–57), the male to female ratio 1.56 (648:415). The median observation time was 57 months (range: 12–206). Five hundred sixty-seven (53.5%) patients were treated for acute lymphoblastic leukemia (ALL), 490 (46%) for acute myeloid leukemia (AML), and 6 (0.5%) for an undifferentiated AL. Type of disease and BMT In the AML group (n 5 490), 426 (87%) patients were grafted in CR1, and 64 (13%) in CR2. In the ALL group (n 5 567), 357 patients (63%) were grafted in CR1 and 210 (37%) in CR2. Autologous and allogeneic BMT were respectively performed in 375 (35%) and 688 (65%) patients. Among the 688 allografted patients, the majority (n 5 658) had a genotypically identical sibling donor marrow. Conditioning chemotherapy Several protocols of high-dose chemotherapy were given before or after TBI prior to BMT. Chemotherapy consisted of cyclophosphamide (Cy) alone in 724 (68%) patients, or a combination without Cy in 317 (30%) patients including 68 pediatric patients who received a combination of melphalan and aracytine. A high dose of only one drug was administered in 22 (2%) patients. Prophylactic medical treatments To prevent veno-occlusive disease (VOD), 195 (18%) patients received heparin at a dose of 100 IU/kg/day during at least 30 days following BMT.

Volume 41, Number 3, 1998

Table 1. Characteristics of patients Age Mean Median Gender Male Female Type of disease ALL AML AUL Remission status CR1 CR2 Conditioning chemotherapy Cy120 Cy200 Combination without Cy* Other drugs Type of BMT Autologous Allogeneic geno.id. Other allogeneic BMT Prophylactic medical treatments Steroids† Administered No Unknown Heparin (VOD prevention) Administered No Unknown Radiotherapy parameters Single-dose TBI Median dose (Gy) Median dose rate (Gy/min) Fractionated TBI Median dose (Gy) Median dose rate (Gy/min) Median number of fractions Median cranial dose†† (n 5 52)

25 6 12 23 (Range: 1–57) 648 (61%) 415 (39%) 567 (53.3%) 490 (46%) 6 (0.7%) 788 (74.2%) 275 (25.8%) 708 (66.5%) 16 (1.5%) 317 (30%) 22 (2%) 375 (35%) 658 (62%) 30 (3%) 316 (30%) 663 (62%) 84 (8%) 195 (18%) 848 (80%) 20 (2%) 10 (6–11.82) 0.06 (0.018–0.3) 12 (8.5–16) 0.065 (0.02–0.56) 6 (2–12) 18 Gy (4–25)

Abbreviations: ALL: acute lymphoblastic leukemia; AUL: acute undifferentiated leukemia; AML: acute myeloblastic leukemia; CR1: first complete remission; Cy: cyclophosphamide (Cy120: 60 mg/kg, 2 consecutive days; Cy200: 50 mg/kg, 4 consecutive days); BMT: bone marrow transplantation; Geno.id.: genotypically identical sibling donor; TBI: total body irradiation; VOD: veno-occlusive disease. * Including 68 pediatric patients who received aracytine and melphalan. † Steroids administered for any reason at a dose of 1 mg/kg/d. †† CNS prophylaxis in ALL patients

The administration of steroids whatever the reason at a dose of 1 mg/kg/day was registered in 316 (30%) patients, and its duration was precised in 228 of 316 patients. Steroid therapy was administered during a short period of less than 1 month in 164, over 1 to 3 months in 45, or for a longer period (. 100 days) in 19 patients. Cranial prophylactic irradiation was administered to 52 (5%) patients. The median dose was 18 Gy (mean: 17.6 6 5.8, range: 4 –25 Gy).

Cataracts after total body irradiation in acute leukemia

Systemic radiotherapy The details of different TBI parameters are presented in Table 1. Single-dose TBI (STBI) was delivered in 495 (46%) patients. A median total dose of 10 Gy (mean: 9.00 6 1.20, range: 6.00 –11.82 Gy) was administered at a median dose rate (DR) of 0.06 Gy/min (mean: 0.11 6 0.096, range: 0.018 – 0.300 Gy/min). In the STBI group, a Co-60 teletherapy unit was used in 246 (23%) patients, and a 4 –25 MV-linear accelerator in 249 (23%) cases. Fractionated TBI (FTBI) was administered to 568 (54%) patients. The median dose was 12 Gy (mean dose: 11.8 6 1.3, range: 8.5–16 Gy) in 2 to 12 fractions. More than half of the patients (333 of 568, 59%) were treated in 6 fractions. The median DR was 0.065 Gy/min (mean: 0.09 6 0.007, range: 0.02– 0.56). A Co-60 teletherapy unit was used in 195 (18%) patients, and a 4- to 25-MV linear accelerator in 373 (36%). The median observation time in the STBI and in the FTBI groups was 74 (12–206) and 49 (12–166) months respectively (p , 1024). Dose rate groups The limits of the DR groups were fixed using the Byar and Peto methods (7, 8) in determination of continuous variables. The cutoff point was 0.04 Gy/min. The analysis concerned two groups treated with low DR # 0.04 Gy/min (LDR, n 5 311, 29%) or high DR . 0.04 Gy/min (HDR, n 5 747, 71%). DR was not registered in 5 cases (one in STBI, and 4 in FTBI group). In the STBI group, 178 (36%) patients were treated by LDR and 316 (64%) by HDR. In the FTBI group, the distribution of DR was respectively 114 (23%) and 387 (77%). The median DR was 0.06 (0.018 – 0.300) Gy/min in the STBI, and 0.065 (0.020 – 0.560) Gy/min in the FTBI groups (p 5 0.061). The median observation time in the LDR and HDR groups were 53 (12–164) and 59 (12–206) months, respectively (p 5 0.59). Statistical methods Means were compared by the Student’s t test. Proportions were compared using the chi-square test for values greater than 5, and Fisher’s exact test for those less than or equal to 5. The probabilities of cataract incidence and cataract surgery were calculated according to the method of Kaplan and Meier (9). Confidence intervals were calculated from standard errors. Because of the heterogeneity in terms of grading of cataract formation by different teams, the analyses were done separately for both events: diagnosis of cataract and surgery for cataract. Patients who died without having cataracts were censored at time of death. The logrank test was used to compare different patient groups (8). The Cox stepwise-regression analysis (10) was performed to estimate the independent effect of potential risk factors. All variables associated with a p-value less than 0.2 in univariate analyses were included in the multivariate model. For all comparisons, quantitative variables were categorized into two or three groups. Each variable was first divided into five cat-

●

Y. BELKACE´ MI et al.

661

egories at approximatively the 20th, 40th, 60th, and 80th percentiles. If the relative rates in two or more adjacent categories were not substantially different, these categories were grouped together (7, 8). If no clear pattern was observed, the median was taken as a cutoff point. RESULTS After a median observation time of 57 (12–206) months, 257 of 1063 (24%) patients developed a cataract; it was bilateral in 239 (22.3%), and unilateral in 18 (1.7%). The overall 5- and 10-year estimated cataract incidence (ECI) in all patients was 27% (6 2) and 50% (6 3), respectively. The prognostic factors influencing cataract formation in univariate analyses are presented in Table 2. In terms of age, patients older than 23 years had a tendency to develop more cataracts but the difference was not statistically significant (p 5 0.08). No difference was observed according to the gender (p 5 0.54). In Figs. 1 and 2 are presented the 5-year probabilities of cataract by age in single and FTBI patients. Influence of radiotherapy parameters According to the type of fractionation, a cataract was observed in 175 of 495 (35%) patients in the STBI group, and 82 of 568 (14%) patients in the FTBI group. The 10-year ECI was 60% (6 3) in the STBI group compared to 43% (6 6) in the FTBI group # 6 fractions, and 7% (6 5) in the FTBI group . 6 fractions (HFTBI), (p , 1024) (Fig. 3). The analysis of the TBI dose showed a trend toward an increased ECI in patients who received a STBI dose . 8 Gy, compared to those who had a dose # 8 Gy (p 5 0.086). In the FTBI group, no difference was observed between patients who received less than 12 Gy and those who received $ 12 Gy (p 5 0.48). The 10-year ECI was significantly lower in the LDR (30% [6 5]) group than in the HDR (59% [6 4]) groups (p , 1024) (Fig. 4). In Figs. 5 and 6 are presented the 5-year probabilities of cataract by dose rates in single and FTBI patients. Moreover, the cataract risk according to the TBI dose is presented in the two groups in Figs. 7 and 8. Overall, the cataract incidence was not significantly higher in the 52 patients who had prophylactic cranial irradiation (p 5 0.31), but patients who received more than 18 Gy had a higher 10-year ECI of 68% (6 13) than those who received a dose # 18 Gy, 36% (6 12), (p 5 0.01). Type of BMT and graft-vs.-host disease (GvHD) Among the 658 allogeneic patients who received an Human Leukocyte Antigen (HLA) identical sibling donor marrow, 188 (28.5%) developed a cataract compared to 60 of 375 (16%) autologous patients. The 10-year ECI was 65% (6 3) in the allogeneic BMT group vs. 46% (6 4) in the autologous BMT group (p 5 0.0018) (Fig. 9). In allogeneic marrow recipients, we did not find any relationship between

662

I. J. Radiation Oncology

●

Biology

●

Physics

Volume 41, Number 3, 1998

Table 2. Prognostic factors in cataract formation and the need for surgery. univariate analyses* Factors Age # 23 . 23 Gender Male Female Type of BMT Auto. Allo.id. cGvHD No Yes Duration of steroids (d)† , 30 30–100 . 100 Use of heparin No Yes Fractionation STBI FTBI (# 6 fr.) HFTBI (. 6 fr.) Total TBI dose (Gy) STBI # 8 .8 FTBI , 12 $ 12 Dose rate (Gy/min) In all patients†† # 0.04 . 0.04 In STBI # 0.04 . 0.04 In FTBI # 0.04 . 0.04 CNS total dose (Gy) # 18 . 18

Number of patients

Cataracts/Surgery

10-yr ECI % CI

p-Value

10-yr PCS % CI

534 529

112/42 142/67

46 (4) 54 (4)

0.08

14 (2) 27 (3)

0.026

648 415

154/68 103/43

33 (4) 47 (4)

0.54

23 (3) 19 (3)

0.86

375 658

60/24 188/80

46 (4) 65 (3)

0.0018

15 (3) 23 (3)

0.0501

412 276

111/44 86/43

54 (5) 56 (5)

0.54

21 (3) 29 (5)

0.085

164 45 19

34/12 14/3 13/9

27 (4) 39 (9) 73 (10)

848 195

225/93 31/17

53 (3) 33 (6)

495 505 63

175/95 80/16 2/0

60 (3) 43 (6) 7 (5)

400 95 463 105

136/68 39/27 70/14 12/2

58 (4) 63 (7) 42 (6) 35 (15)

311 747

37/19 219/92

30 (5) 59 (4)

178 316

32/19 143/76

42 (6) 69 (4)

114 387

5/0 74/16

7 (5) 55 (7)

32 18

6/3 10/7

36 (12) 68 (13)

,1024 0.042

,1024 0.086

p-Value

—

—

22 (2) 18 (4)

0.90

32 (3) 3 (2) 0

,1024

28 (3) 47 (3) 7 (2) 5 (3)

0.0028

,1024

12 (3) 25 (3)

0.0033

,1024

19 (5) 38 (4)

0.0055

0 10 (3)

0.015

11 (7) 54 (16)

0.07

0.48

,1024 0.01

0.70

Abbreviations: ECI: estimated cataract incidence; PCS: probability of cataract surgery; CI: confidence interval; BMT: bone marrow transplantation; Auto: autologous BMT; Allo.id.: allogeneic BMT (genotypically identical sibling-donor marrow); cGvHD: chronic graft-vs.-host disease; STBI: single-dose total body irradiation; FTBI: fractionated TBI; HFTBI: hyperfractionated TBI (FTBI . 6 fractions); CNS: central nervous system. * Kaplan-Meier and logrank test. † Applicable only for 5-yr probabilities. †† Dose rate not registered in 5 cases (one in STBI and 4 in FTBI).

cataract formation and development of acute (p 5 0.61) or chronic graft-vs.-host disease (cGvHD) (p 5 0.54).

VOD. The 10-year ECI was 33% (6 6) in the first group, and 53% (6 3) in the second group (p 5 0.042) (Fig. 10).

Associated medical treatments The 5-year ECI increased with the duration of steroids; for more than 100 days the 5-year ECI was 73% (6 10), as compared to 39% (6 9), and 27% (6 4), after a duration of 30 –100 and , 30 days respectively (p , 1024). In the 195 patients who received heparin for VOD prophylaxis, 31 (16%) developed a cataract compared to 225 of 848 (26.5%) patients who had no preventive treatment for

Multivariate analysis Table 3 lists the variables associated with a significant relative risk (RR) (positive or negative) of cataractogenesis in the multivariate analysis. In all patients, the variables that were independently associated with an increased risk of cataract were: (1) age . 23 years (RR 5 1.34), (2) DR of TBI . 0.04 Gy/min (RR 5 3.26), (3) allogeneic BMT (RR 5 1.65), and (4) administration of steroids during a

Cataracts after total body irradiation in acute leukemia

●

Y. BELKACE´ MI et al.

663

Fig. 1. Cataract risk by age in single dose TBI.

period . 100 days (RR 5 1.36). Conversely, the use of FTBI regimen was associated with a decreased risk of cataract (RR 5 0.36). Multivariate models were also constructed to determine the risk factors separately in the STBI and FTBI groups. In patients who received STBI, a very high risk of cataract development was associated with a dose rate . 0.04 Gy/min (RR 5 3.26), the administration of steroids . 100 days (RR 5 1.83), and a STBI dose exceeding 8 Gy (RR 5 1.98); while heparin administration was a protective factor (RR 5 0.4). In the FTBI patients, DR . 0.04 Gy/min and allogeneic BMT were associated with higher risk of cataract formation (RR 5 4.34 and 2.08, respectively). Risk factors for cataract surgery One hundred eleven (43%) of the 257 patients with cataracts required surgery after a median period of 9 months (range: 1–79). Cataract surgery was performed by extracapsular extraction and intraocular lens implantation in the majority of patients (109 of 111; 98%); whereas the technique was not given in two cases (2%). The prognostic factors by univariate analyses for requiring cataract surgery in all patients undergoing TBI and BMT are presented in Table 3. The overall rate and the probability for requiring surgical cataract removal were significantly higher after STBI than after FTBI # 6 fractions (p , 1024). In the STBI group, a total dose . 8 Gy was associated with a higher probability for surgery (p 5 0.0028). In all patients, as well as in the STBI or FTBI (# 6 fractions) groups, patients treated with a DR . 0.04 Gy/min required significantly more cataract surgery than those treated with a DR # 0.04 Gy/min. In addition, older age (. 23 years; p 5 0.026) and a long-term steroid

Fig. 2. Cataract risk by age in fractionated TBI.

Fig. 3. Product-limit estimates for cataract incidence according to fractionation. STBI: single-dose total body irradiation; FTBI: fractionated TBI (# 6 fractions); HFTBI: hyperfractionated TBI (. 6 fractions).

administration (. 100 days; p , 1024) were also associated with a higher probability for requiring cataract surgery. Multivariate analyses were performed in all patients and in the STBI group. In the FTBI group, the number of events was not sufficient to construct a multivariate model. When considering the whole population, the factors independently associated with an increased need for cataract surgery were an older age (. 23 years; RR 5 1.7) and a dose rate . 0.04 Gy/min (RR 5 2.06) whereas fractionation was associated with a decreased risk (RR 5 0.14). Although a protective effect of heparin (RR 5 0.43) and the development of cGvHD (RR 5 2.47) were not significant factors on the univariate analyses, they became significant once adjustment was made for other factors in the multivariate analysis. In the STBI group, a total dose . 8 Gy (RR 5 2.48) and a DR . 0.04 Gy/min (RR 5 2.06) were again significant risk factors. Moreover, the administration of heparin significantly decreased the risk for requiring cataract surgery (RR 5 0.26), while cGvHD increased it (RR 5 2.12). Outcome after cataract surgery Following cataract surgery, the ophthalmologic follow-up did not exceed 1 year in 10 of 111 (9%) patients.

Fig. 4. Product-limit estimates for cataract incidence according to fractionation and dose rate. STBI: single-dose total body irradiation; FTBI: fractionated TBI; LDR: low dose rate (# 0.04 Gy/ min); HDR: high dose rate (. 0.04 Gy/min).

664

I. J. Radiation Oncology

●

Biology

●

Physics

Volume 41, Number 3, 1998

Fig. 5. Cataract risk by dose rate in single dose TBI.

The median follow-up period in the remaining 101 operated patients was 71 months (range: 12–151). To evaluate secondary posterior capsular opacification, we analyzed the results separately in both eyes. Among the 101 operated left eyes, 15 (14.9%) developed secondary capsular opacification. Following right eye surgery (n 5 92), secondary capsular opacification was observed in 12 (13%) patients. YAG (yttrium aluminum garnet)-laser capsulotomy was required in 6% of patients. Among the 89 patients who had an ophthalmologic follow-up $ 2 years after surgery, the overall secondary capsular opacification rate was 15.7% (14/89). DISCUSSION The mechanism of cataract formation after irradiation is not completely understood. Abnormal fibers migrate toward the posterior pole or central region of the lens where they generate multiple subcapsular opacities (11–15). The relation between cataract formation and exposure to ionizing radiation has been known since 1897 (16). In animal models, radiation-induced cataract has been reported after exposure to different type of ionizing radiation (17). In humans, an increased risk of cataractogenesis was found in the late forties in a population of cyclotron workers and survivors of atomic bombs (18, 19), and later in patients treated by local radiotherapy for head and neck cancers (20). Merriam and Focht (21, 22) were the first to describe the relationship between total dose, rate of exposure, and cataractogenesis. The role of TBI in cataract development after BMT is well known. However, the overall rates, severity, and the need for surgery depend on several factors. In patients given only chemotherapy before BMT, cataract rates (0 –19%) are lower than patients conditioned with TBI (3, 23–26). In the

Fig. 6. Cataract risk by dose rate in fractionated TBI.

Fig. 7. Cataract risk by dose in single dose TBI.

Seattle experience (23), a lower risk of cataract was noted in recipients of chemotherapy (19%) compared to 12 Gy FTBI patients (34%). Others (5, 24) have reported in children and adults an increased rates of cataracts in recipients of FTBI (63– 83%) as compared to no TBI patients (5–9%). The type of fractionation is important; indeed in STBI regimens, several studies have reported overall cataract rates ranging from 11 to 100% (3, 5, 25–34), whereas in FTBI schemes the overall cataract incidence ranged from 0 to 83%, after a follow-up period of 10 months to 11 years (3, 5, 23–25, 27, 33, 34). In this report, the overall cataract rates were 35% in the STBI group and 14% in the FTBI (# 6 fractions) group after a median observation time of 74 and 49 months, respectively. Numerous reports (5, 23, 25, 26, 29, 33) showed a sparing effect of fractionated regimens as compared to STBI. As in this study, the severity of progressive lens opacification and the probability of requiring surgery were significantly greater after STBI (23, 26, 29). Besides fractionation, previous reports have shown the influence of dose rate on cataract formation, and the risk for requiring cataract surgery both in STBI and FTBI regimens (5, 24, 29, 33). However, when the number of fractions is higher than 6, the importance of DR disappears. In the current report, the ‘‘threshold’’ for high risk of cataract

Fig. 8. Cataract risk by dose in fractionated TBI.

Cataracts after total body irradiation in acute leukemia

Fig. 9. Product-limit estimates for cataract incidence according to the type of bone marrow transplantation (Allogeneic from identical sibling donnor marrow [Allo HLA id.] vs. Autologous transplant [auto.]).

formation in the STBI and FTBI (# 6 fractions) groups was 0.04 Gy/min. The 10-year ECI and the probability for requiring cataract surgery were significantly increased in patients treated with DR . 0.04 Gy/min (Fig. 4). In the Hoˆpital Tenon experience, both from a randomized study of DR (33), and a retrospective analysis of 494 patients (5), the influence of DR has been primordial overcoming: none of the 36 STBI patients treated by a low dose rate (, 0.048 Gy/min) has developed a cataract. The cataractogenic property of steroids, their dose-effect, and duration have been reported in rheumatoid arthritis and many other diseases in the early 1960s (35–39). After BMT, the role of steroids in cataractogenesis is difficult to isolate from the other powerful factors. Nevertheless, the effect of ionizing radiation and steroids seems to be additive. Indeed, patients who had STBI and steroids had an increased cataract relative risk of 19 to 21 times compared to those given neither TBI nor steroids (25, 26). In this study, the risk of cataract formation increased with the duration of steroid therapy after STBI but not after FTBI, however the duration of steroid therapy was registered in only 21% of patients. In contrast, the Seattle group, studying more finely the influence of steroids, have noted that a 100-day administration of steroids increased cataract development in FTBI and no TBI patients, whereas its importance disappeared after STBI regimen (23). Other reports (5, 24, 34, 37) have not noted any significant influence of steroids on cataractogenesis whatever the TBI regimen used in children or adults. This may be related to the small population of patients, or to the interaction with other factors such as graft-vs.-host disease (GvHD) or allogeneic BMT. Indeed, the influence of GvHD and posttransplant immunosuppression on lens opacification is extremely difficult to sort out in such a complex biological setting. This suggests that any apparent influence of these factors on the cataract formation is a reflection of the close relationship between steroid administration and GvHD. We did not observe in this study any relationship between GvHD and cataract development. However, in terms of the

●

Y. BELKACE´ MI et al.

665

need for surgery, multivariate analysis showed that higher risk for requiring cataract surgery is associated with cGvHD rather than with long-term steroid therapy. The crystalline lens flexibility decreases after 40 years of age, and its opacification appears later. In BMT patients, the influence of age is controversial. The overall reported cataract rates in pediatric series are relatively high, 32% to 80% (3, 24, 28, 37, 40, 41). Several reports did not find any influence of age (5, 25, 26, 41). In this study, age . 23 years which was not a significant factor in univariate analysis, became significant once adjustment was made for other factors in multivariate analysis. On the other hand, indications for cataract surgery vary between centers, and are not always the same for adults and children. Early surgery in children may be necessary to prevent amblyopia, while later in childhood one is inclined to postpone cataract surgery to preserve accommodation for as long as possible. This is probably the rationale for the reported low risk for cataract surgery in patients under 25 years (29). Indeed, to date the reported data cannot formally demonstrate whether or not a young lens is more susceptible to the effects of irradiation. As reported by others (5, 25, 26, 29), gender was not identified as a factor influencing the cataract formation or the need for cataract surgery. However, in series concentrating only on children, severity and cataract rates are observed more frequently in girls (37) or in boys (40), with sometimes a slightly earlier detection in the latter (42). Cranial irradiation has been an effective mean of eliminating subclinical arachnoid leukemic infiltration in children with ALL (43), and 13–20% of the cranial dose is received by the lens (44). However, the influence of cranial radiotherapy boost on cataract formation cannot be easily isolated from other factors such as diagnosis of ALL. Indeed, there is obviously a close correlation between diagnosis of ALL and the likelihood of receiving cranial irradiation (25), which suggests that the risk of cataract is increased by cranial irradiation rather than the type of disease. Moreover, despite the use of lens shielding during STBI, cranial radiotherapy is reported as the most important independent predictive factor for cataract surgery, mostly in patients irradiated during more than 1 year preceeding TBI (29). In this study, we noted a

Fig. 10. Product-limit estimates for cataract incidence according the administration of heparin.

666

I. J. Radiation Oncology

●

Biology

●

Physics

Volume 41, Number 3, 1998

Table 3. Independent risk factors for cataract formation and the need for surgery in all patients, STBI, and FTBI groups. Multivariate analyses* Cataract formation Covariables All patients Age . 23 DR . 0.04 Gy/min FTBI Allogeneic BMT Use of steroids . 100 d Use of heparin cGvHD STBI group DR . 0.04 Gy/min Use of steroids Use of heparin TBI dose . 8 Gy cGvHD FTBI group† DR . 0.04 Gy/min Allogeneic BMT

Cataract surgery

RR (CI)

p-Value

RR (CI)

p-Value

1.34 (1.02–1.75) 3.26 (2.26–4.72) 0.36 (0.27–0.48) 1.65 (1.18–2.30) 1.36 (1.01–1.83)

0.033 ,1024 ,1024 0.0032 0.044

1.7 (1.12–2.58) 2.06 (1.23–3.45) 0.14 (0.08–0.25)

0.013 0.0059 ,1024

0.43 (0.24–0.78) 2.47 (1.63–3.74)

0.0058 ,1024

,1024 0.0006 0.0007 0.0008

3.26 (2.15–4.93) 1.83 (1.29–2.58) 0.40 (0.23–0.68) 1.98 (1.33–2.95) 4.34 (1.74–10.81) 2.08 (1.16–3.71)

2.06 (1.2–3.52)

0.0087

0.26 (0.14–0.51) 2.48 (1.47–4.18) 2.12 (1.34–3.37)

0.0001 0.0006 0.0014

0.0016 0.014

Abbreviations: RR: relative risk; CI: confidence interval; DR: dose rate; FTBI: fractionated total body irradiation; cGVHD: chronic graft-versus-host disease; STBI: single-dose TBI. * Proportional hazards regression model (Cox model). † Number of events was not sufficient to perform a multivariate analysis in the FTBI group.

dose-effect (. 18 Gy) of cranial radiotherapy; whereas in some other reports, cranial irradiation had no influence on cataract formation (5, 33, 42). As reported recently (5), this study reveals not only the protective effect of heparin against cataract formation after STBI but also its impact on decreasing the need for cataract surgery. This observation is intriguing as the crystalline lens has no blood supply. Nevertheless, an inhibiting effect of heparin on the proliferation of epithelial lens cells has been reported in a bovine model of in vitro cultures (45). In senile cataract population, the use of heparin-surface-modified lens for intraocular implantation decreases significantly the inflammation and the risk of secondary cataract following surgery (46). Furthermore, the in vivo cell proliferation from the germinative zone of the capsular bag following ECS and lens implantation seems to be heparin-dose dependent (47). In senile cataract, following extracapsular surgery and lens implantation, secondary opacification rises in about 50 –70% of patients 2–3 years following surgery whatever the technique used (48, 49). Radiation-inhibiting effect on secondary proliferation after ECS was suggested from an

experimental model (6). In a recent report concerning TBI and BMT, only 10% of the operated patients developed a secondary cataract (5). In the current report, among the 89 patients operated and followed for at least 2 years, 14 patients (15.7%) had a secondary capsular cataract. Secondary opacification following ECS and lens implantation implies epithelial cell proliferation. Nevertheless, the lower rates of secondary cataract among the irradiated patients could be related to a better ‘‘tolerance’’ of the implanted lens, certainly due rather to an antiproliferative and/or a local immunosuppressive effect of ionizing radiation than to their younger age than the senile population. Indeed, it is known that younger persons have a higher tendency to develop secondary cataracts (50). In conclusion, this retrospective multicenter study realized in a large number of patients who received TBI for BMT has confirmed the influence of the TBI technique on cataractogenesis. Heparin, used in the prevention of VOD, has a protective effect. Whether, on the other hand, irradiation prevents secondary capsular opacification following ECS and lens implantation remains to be studied.

REFERENCES 1. Barrett, A. Total body irradiation before bone marrow transplantation: A review. Clin. Radiol. 33:131–135; 1982. 2. Brauner, R.; Fontoura, M.; Zucker, J. M. Growth and growth hormone secretion after bone marrow transplantation. Arch. Dis. Child. 68:458 – 463; 1993. 3. Giri, N.; Davis, E. A. C.; Vowels, M. R. Long-term compli-

cations following bone marrow transplantation in children. J. Paediatr. Child Health 29:201–205; 1993. 4. Kim, T. H.; McGlave, P. B.; Ramsay, N.; Woods, W.; Bostrom, B.; Vercellotti, G.; Hurd, D.; Krivit, W.; Nesbit, M.; Haake, R.; Khan, F.; Kersey, J. H. Comparison of two total body irradiation regimens in allogeneic bone marrow trans-

Cataracts after total body irradiation in acute leukemia

5.

6.

7.

8.

9.

10. 11. 12. 13. 14. 15. 16. 17.

18. 19. 20. 21. 22.

23.

24.

plantation for acute non-lymphoblastic leukemia in first remission. Int. J. Radiat. Oncol. Biol. Phys. 19:889 – 897; 1990. Belkace´mi, Y.; Ozsahin, M.; Pe`ne, F.; Rio, B.; Laporte, J. P.; Touboul, E.; Leblond, V.; Schlienger, M.; Gorin, N. C.; Laugier, A. Cataractogenesis after total body irradiation. Int. J. Radiat. Oncol. Biol. Phys. 35:53– 60; 1996. Roy, F. H.; Olmos, E. After cataracts: Studies of chemical and radiation inhibition. Contact Intraoc. Lens Med. J. 5:175–178; 1979. Byar, D. P. Identification of prognostic factors. In: Buyse, M. E., Staquet, M. J., Sylvester, M. J., eds. Cancer clinical trials methods and practice. London, Oxford: Medical Publications; 1988:423. Peto, R.; Pike, M. C.; Armitage, P.; Breslow, M. E.; Cox, D. R.; Howard, S. W.; Mantel, N.; McPherson, K.; Peto, J.; Smith, P. G. Design and analysis of randomised trials requiring prolonged observation of each patient: Part II. Br. J. Cancer 35:1–39; 1977. Kaplan, E. L.; Meier, P. Non parametric estimation from incomplete observations. J. Am. Stat. Assoc. 53:457– 481; 1958. Cox, D. R. Regression models and life tables. J. R. Stat. Soc. 34:187–202; 1972. Hall, E. J. Radiation cataractogenesis. In: Hall, E. J., eds. Radiobiology for the radiologist. Philadelphia: Lippincott; 1994:378 –384. Merriam, G. R.; Worgul, B. Experimental radiation cataract: Its clinical relevance. New York: Bull. Acad. Med. 59:372– 392; 1983. Sperduto, R.; Hiller, R. The presence of nuclear, cortical and posterior subcapsular lens opacities in a general population sample. Ophthalmology 91:815– 818; 1984. Von Sallman, L. The lens epithelium in the pathogenesis of cataract. Am. J. Ophthalmol. 44:1570 –1598; 1957. Von Sallman, L. Early lenticular lesions resulting from ionising radiation. Trans. Am. Acad. Ophthalmol. 63:439 – 448; 1959. Chalupecky, H. Ueber die Wirkung der Ro¨ntgenstrahlen auf das Auge und die. Haut. Zentralbl. F. Prak. Augenh. 21:234 – 267; 1897. Darden, E. B. Jr; Christenberry, K. W.; Beauchamp, J. J.; Bender, R. S.; Jernigan, M. C.; Conklin, J. W.; Upton, A. C. Comparison of 60-MeV protons and 300-kVp x-rays for induction of lens opacities in RF mice. Radiat. Res. 43:598 – 601; 1970. Abelson, R.; Kruger, P. Cyclotron induced cataracts. Science 110:655– 656; 1949. Cogan, D.; Martin, S.; Kimura, S. Atom bomb cataracts. Science 110:654 – 655; 1949. Macfaul, P. A.; Bedford, M. A. Ocular complications after therapeutic irradiation. Br. J. Ophthalmol. 54:237–247; 1970. Merriam, G. R.; Focht, E. F. A clinical study of radiation cataracts and the relationship to dose. Am. J. Roentgenol. 77:759 –785; 1957. Merriam, G. R.; Focht, E. F. A clinical and experimental study of the effect of single and divided doses of radiation on cataract production. Trans. Am. Ophthalmol. Soc. 60:35–52; 1962. Benyunes, M. C.; Sullivan, K. M.; Deeg, J. H.; Mori, M.; Meyer, W.; Fisher, L.; Bensinger, R.; Jack, M. K.; Hicks, J.; Witherspoon, R.; Buckner, C. D.; Hansen, J. A.; Appelbaum, F. R.; Storb, R. Cataracts after bone marrow transplantation: Longterm follow-up of adults treated with fractionated total body irradiation. Int. J. Radiat. Oncol. Biol. Phys. 32:661– 670; 1995. Bray, L. C.; Carey, P. J.; Proctor, S. J.; Evans, R. G. B.;

25.

26.

27. 28.

29.

30.

31. 32. 33.

34.

35.

36. 37. 38.

39. 40. 41.

42.

●

Y. BELKACE´ MI et al.

667

Hamilton, P. J. Ocular complications of bone marrow transplantation. Br. J. Ophthalmol. 75:611– 614; 1991. Deeg, H. G.; Flournoy, N.; Sullivan, K. M.; Sheehan, B.; Buckner, C. D.; Sanders, J. E.; Storb, R.; Witherspoon, R. P.; Thomas, E. D. Cataracts after total body irradiation and marrow transplantation: A sparing effect of dose fractionation. Int. J. Radiat. Oncol. Biol. Phys. 10:957–964; 1984. Tichelli, A.; Gratwohl, A.; Egger, T.; Roth, J.; Prunte, A.; Nissen, C.; Speck, B. Cataract formation after bone marrow transplantation. Ann. Intern. Med. 119:1175–1180; 1993. Barrett, A.; Nicholls, J.; Gibson, B. Late effects of total body irradiation. Radiother. Oncol. 9:131–135; 1987. Calisendorff, B. M.; Bolme, P.; El Azazi, M. The development of cataract in children as late side-effect of bone marrow transplantation. Bone Marrow Transplant. 7:427– 429; 1991. Fife, K.; Milan, S.; Westbrook, K.; Powels, R.; Tait, D. Risk factors for requiring cataract surgery following total body irradiation. Radiother. Oncol. 33:93–98; 1994. Hamon, M. D.; Gale, R. F.; Macdonald, I. D.; Smith, O. P.; Collins, C. H.; Skeggs, D. B.; Gandhi, L.; Prentice, H. G. Incidence of cataracts after single fraction total body irradiation: The role of steroids and graft versus host disease. Bone Marrow Transplant. 12:233–236; 1993. Lappi, M.; Rajantie, J.; Uusitalo, R. J. Irradiation of cataract in children after bone marrow transplantation. Graefes Arch. Clin. Exp. Ophthalmol. 228:218 –221; 1990. Liversey, S. J.; Holmes, J. A.; Wittaker, J. A. Occular complications of bone marrow transplantation. Eye 3:271–276; 1989. Ozsahin, M.; Belkace´mi, Y.; Pe`ne, F.; Dominique, C.; Schwartz, L. H.; Uzal, C.; Lefkopoulos, D.; Gindrey-Vie, B.; Vitu-Loas, L.; Touboul, E.; Schlienger, M.; Laugier, A. Total body irradiation and cataract incidence: A randomized comparison of two instantaneous dose rates. Int. J. Radiat. Oncol. Biol. Phys. 28:343–347; 1994. Ozsahin, M.; Pe`ne, F.; Touboul, E.; Gindrey-Vie, B.; Dominique, C.; Lefkopoulos, D.; Krzisch, C.; Balosso, J.; Vitu, L.; Schwartz, L. H.; Rio, B.; Gorin, N. C.; Leblond, V.; Schlienger, M.; Laugier, A. Total-body irradiation before bone marrow transplantation: Results of two randomized instantaneous dose rates in 157 patients. Cancer 69: 2853–2865; 1992. Black, R. L.; Oglesby, R. B.; Von Sallmann, L.; Bunim, J. J. Posterior subcapsular cataracts induced by corticosteroids in patients with rheumatoid arthritis. JAMA 174:166 – 171; 1960. Crews, S. J. Posterior subcapsular lens opacities in patients on long-term corticosteroid therapy. Br. Med. J. i:1644 –1647; 1963. Hoover, D. L.; Smith, L. E. H.; Turner, J. S.; Gelber, L. D.; Sallan, S. E. Ophthalmic evaluation of survivors of acute lymphoblastic leukemia. Ophthalmology 95:151–155; 1988. Kobayashi, Y.; Akaishi, K.; Nishio, T.; Kobayashi, Y.; Kimura, Y.; Nagata, M. Posterior subcapsular cataract in nephrotic children receiving steroid therapy. Am. J. Dis. Child. 128:671– 673; 1974. Urban, R. L.; Cotlier, E. Corticosteroid-induced cataracts. Surg. Ophthalmol. 31:102–110; 1986. Elliott, A. J.; Oakhill, A.; Goodman, S. Cataract in childhood leukemia. Br. J. Ophthalmol. 69:459 – 461; 1985. Schmidt, G.; Niland, J. C.; Forman, S. J.; Fonbuena, P. P.; Dagis, A. C.; Grant, M. M.; Ferrell, B. R.; Barr, A. T.; Stallbaum, B. A.; Chao, N. J.; Blum, K. G. Extended follow-up in 212 patients long-term allogeneic bone marrow transplant survivors. Transplantation 55:551–557; 1993. Calisendorff, B. M.; Bolme, P. Cataract development and

668

43.

44. 45. 46.

I. J. Radiation Oncology

●

Biology

●

Physics

outcome of surgery in bone marrow transplanted children. Br. J. Ophthalmol. 77:36 –38; 1993. Inati, A.; Sallan, S. E.; Cassady, J. R.; Hitchcock-Bryan, S.; Clavell, L. A.; Belli, J. A.; Sollee, N. Efficacy and morbidity of central nervous system ‘‘prophylaxis’’ in childhood acute lymphoblastic leukemia: Eight years experience with cranial irradiation and intrathecal methotrexate. Blood 61:297–303; 1983. Kline, R. W.; Gillin, M. T.; Kun, L. E. Cranial irradiation in acute leukemia dose estimate in the lens. Int. J. Radiat. Oncol. Biol. Phys. 5:117–121; 1979. Knor, M.; Wunderlich, K.; Shenhl, K. P.; Thiel, H. J.; Dartsh, P. C. Effect of heparin on proliferation of cultivated bovine lens epithelial cells. Ophthalmology 89:319 –324; 1992. Borgioli, M.; Coster, D. J.; Fan, R. F.; Henderson, J.; Jacobi, K. W.; Kirkby, G. R.; Lai, Y. K.; Menezo, J. L.; Moutard, M.; Strobel, J. Effect of heparin surface modification of polymethylmethacrylate intraocular

Volume 41, Number 3, 1998

47.

48.

49.

50.

lenses on signs of postoperative inflammation after extracapsular cataract extraction: One year results of a doublemasked multicenter study. Ophthalmology 99:1248 –1255; 1992. Philipson, B.; Fagerholm, P.; Calel, B.; Grunge, A.; Hallna¨s, K.; Lydahl, E.; Ohman, L. Heparin surface modified intraocular lens: A one year follow-up of a safety study. Acta Ophthalmol. 68:601– 603; 1990. Kraff, M. C.; Sanders, D. R.; Lieberman, H. L. Total cataract extraction through a 3-mm incision: A report of 650 cases. Ophthal. Surg. 10:46 –54; 1979. Wilhemus, K. R.; Emery, J. M. Posterior capsule opacification following phacoemulsification. Ophthal. Surg. 11:246 –247; 1980. Keech, R. V.; Tongue, A. C.; Williams, E. S. Complications after surgery for congenital and infantile cataracts. Am. J. Ophthalmol. 108:136 –141; 1989.