- Email: [email protected]
Cataractogenesis after cobalt-60 eye plaque radiotherapy
0360-3016/93 $6.00 + .OO Copyright 0 1993 Pergamon Press Ltd.
Inr. J. Radiation Oncology Bml. Phys., Vol. 26, pp. 625-630 Printed in the U.S.A. All rights reserved.
??Clinical Original Contribution
CATARACTOGENESIS MARKUS
KLEINEIDAM,
AFTER COBALT-60 EYE PLAQUE RADIOTHERAPY
M.D.,’ JAMES J. AUGSBURGER, M.D.,’ CARLOS HERNANDEZ, M.D.,2
PATRICK GLENNON, M.S.2 AND LUTHER W. BRADY, M.D.2 ‘Oncology Unit, Retina Service, Wills Eye Hospital, Jefferson Medical College; and %he Department of Radiation Oncology and Nuclear Medicine, Hahnemann University, Philadelphia, Pennsylvania, USA Purpose: This studywas designed to estimate the actuarial incidence of typical postirradiation cataracts and to identifyprognostic factors related to their development in melanoma-containing eyes treated by Cobalt-60 plaque radiotherapy. Our special interest was the impact of calculated radiation dose and dose-rate to the lens. MethodsandMaterial:The authorsevaluated theactuarial occurrence of post-irradiation cataract in 365 patients with primary posterior uveal melanoma treated by Cobalt-60 plaque radiotherapy between 1976 and 1986. Results: Only 22% (S.E. = 4.6%) of the patients who received a total dose of 6 to 20 Gy at the center of the lens developed a visually significant cataract attributable to the radiation within 5 years after treatment. Using multivariate Cox proportional hazards modeling, the authors identified thickness of the tumor, location of the tumor’s anterior margin relative to the equator and the ora serrata, and diameter of the eye plaque used as the best combination of covariables for predicting length of time until development of cataract. Surprisingly, the dose of radiation delivered to the lens, which was strongly correlated to all of these covariables, was not a significant predictive factor in multivariate analysis. Conclusion: The results suggest that success of efforts to decrease the occurrence rate of post-irradiation cataracts by better treatment planning might be limited in patients with posterior uveal melanoma. Uveal melanoma, Cataract, Brachytherapy,
Cobalt-60 eye plaques.
INTRODUCTION
1952 (lo), include posterior subcapsular lens clouding caused by damage to the subcapsular epithelium. In spite of the fact that post-irradiation cataracts are well-known to ophthalmologists, the actuarial incidence of this form of cataract following specific methods of ocular radiotherapy has not been reported. This study was designed to estimate the actuarial incidence of typical postirradiation cataracts and to identify prognostic factors related to their development in melanoma-containing eyes treated by Cobalt-60 plaque radiotherapy. Our special interest was the impact of calculated radiation dose and dose-rate (12) to the lens.
On the basis of currently available, non randomized but statistically-adjusted comparative survival studies, episcleral plaque radiotherapy and enucleation (21) are believed to be nearly equivalent in terms of the survival prognosis of patients having a posterior uveal malignant melanoma ( 1-7, 9, 19). If this is true, plaque treatment has the advantage of preserving vision, at least to some degree, in the tumor-containing eye in a substantial proportion of patients for intervals of up to 5 years or even longer. Unfortunately, the great majority of patients treated by episcleral plaque radiotherapy lose a substantial portion of their vision over 5- 10 years, largely on the basis of radiation-related complications ( 15). One of the most widely recognized and common complications of episcleral plaque brachytherapy in eyes with primary posterior uveal melanoma is cataract formation (8, 14, 15, 17). The classic features of post-irradiation cataract, as described by Cogan, Donaldson and Reese in
Four hundred patients with posterior uveal malignant melanoma were treated primarily by Cobalt-60 plaque radiotherapy at Wills Eye Hospital, Philadelphia, between
Reprint requests to: James J. Augsburger, M.D., Oncology Unit, Retina Service, Wills Eye Hospital, 900 Walnut St., Philadelphia, PA 19 107, USA. Acknowledgement: The authors would like to acknowledge Jerry A. Shields, M.D., who managed some of the patients evaluated in this study.
This study was supported in part by a grant from the Deutsche Krebsgesellschaft, Dr. Mildred Scheel Stiftung, Bonn, Germany, to Dr. Markus Kleineidam, and an unrestricted grant from Research to Prevent Blindness, Inc., New York, NY, U.S.A., to Wills Eye Hospital. Accepted for publication 15 February 1993.
METHODS
AND MATERIALS
Patients
625
1. J. Radiation Oncology 0 Biology 0 Physics
626
1976 and 1986. Eighteen patients received additional plaque radiotherapy for local tumor relapse subsequent to the original radiotherapy and were therefore excluded from this study. The remaining 382 patients (Table 1) included 186 women (48.7%) and 196 (51.3%) men. These patients ranged from 25 to 89 years of age and had a mean age of 59 years. Tumor thickness ranged from 1.5 to 15.5 mm (mean = 6.5 mm, S.D. = 2.4 mm) and largest basal tumor diameter ranged from 5.0 to 22.0 mm (mean = 11.5 mm, S.D. = 3.0 mm). The anterior margin of the tumor was posterior to the equator in 145 eyes, between the equator and ora serrata in 139 eyes and anterior to the ora serrata in 98 eyes. The posterior tumor margin was located within 3 mm of the optic disc in 113 eyes, over 3 mm from the optic disc but posterior to the ocular equator in 246 eyes, and anterior to the equator in 23 eyes. Seventeen of the 382 patients (4.5%) had undergone cataract surgery prior to their plaque radiotherapy and therefore were not at risk to develop a post-irradiation cataract. These 17 patients were excluded prior to final data analysis, leaving us with a study group of 365 patients. Follow-up and endpoint
Patients were followed up until death or their most recent re-examination before the closing date for data analysis (May 1992). Some patients who were unable to return to Wills Eye Hospital for regular periodic followup examination were advised to undergo periodic re-examination and evaluation by their referring ophthalmologist. Two hundred two of the 365 patients (55.3%) had all of their follow-up at Wills Eye Hospital. In contrast, 109 (39.9%) had some of their follow-up at Wills and the rest with their referring ophthalmologist (usually on an
Volume 26, Number 4, I993 visit basis), and 54 (14.8%) had all of their follow-up performed by the referring ophthalmologist. Patients were considered to be lost to follow-up if they had not been re-examined at Wills Eye Hospital or by their local ophthalmologist within 1 year prior to the closing date. For the purposes of this study, we defined a post-irradiation cataract as lens clouding subsequent to radiotherapy which (a) included posterior subcapsular changes as a prominent feature (lo), (b) was of sufficient extent to be a contributing factor to post-irradiation visual impairment, and (c) was asymmetrically more pronounced than any concurrent lens clouding in the fellow eye. No staging of lens clouding was attempted. The endpoint assessed in this study was length of post treatment followup until diagnosis of a post-irradiation cataract.
alternating
Cobalt-60 eye plaque radiotherapy
The Cobalt-60 applicators used were developed by Stallard in 1966 (20). Only plaques of 10 mm (160 cases) and 15 mm (222 cases) were used in this series. Further details have been published previously (5, 15). The radiation doses to the tumor’s apex and base and the overall treatment time were estimated from two-dimensional isodose distributions generated by a computer program (PRIME). The estimated dose-rate 1 mm from the surface of the plaque (base of the tumor) ranged from 0.88 to 2.61 Gy/hr (mean = I .72 Gy/hr, S.D. = 0.27 Gy/ hr). The overall treatment time ranged from 81 to 432 hrs (mean = 211 hrs, S.D. = 55.0 hrs). The basal radiation dose ranged from 171.0 Gy to 702.0 Gy (mean = 355.0 Gy, S.D. = 76.0 Gy) and the apical radiation dose ranged from 26.3 Gy to 157.0 Gy (mean = 83.5 Gy, S.D. = 17.6 Gy).
Table 1. Variables entered in the analysis and their distributions Variable
Mean
S.D.
Range
Age Dose to the Lens Dose-rate at the Lens Duration of Treatment Largest basal Tumor Diameter Thickness of the Tumor
59 years 32.0 Gy 0.15 Gy/h 211 h 11.5 mm 6.5 mm
12.7 18.6 0.09 55.0 3.0 2.4
25-89 years 6-112 Gy 0.04-0.48 Gy/h 81-432 h 5-22 mm 1.5-15.5 mm
Anterior location
1. Posterior to the equator 2. Posterior to the ora serrata 3. Anterior to the ora serrata 1. 3 mm from disc but posterior to the equator 3. Anterior to the equator 10 mm: 160 (41.88%) 15 mm: 222 (58.12%) 196 (51.31%) Male Female 186 (48.69%)
Posterior location
Plaque’s diameter Sex
145 (37.96%) 139 (36.39%) 98 (25.65%) 113 (29.58%) 246 (64.40%) 23 (6.02%)
621
Cataractogenesis after Co-60 plaque radiotherapy 0 M. KLEINEIDAM et al.
;
0.6
model (11). In the multivariate analysis stepwise stepdown regression was carried out, starting with all covariables and excluding one at a time those not associated with a p-value of at least 0.0 1. This final model was then tested against other models incorporating covariables
B $ d
0.4
which were of high predictive analysis.
8 b
0.2
s
0.6
9
E 0 8 !? P
0.0 0
value
in the univariate
RESULTS
2
4
6
6
10
Time after treatment [years]
Fig. 1. Cumulative actuarial rates of developing post-irradiation cataract in 365 patients with primary posterior uveal melanoma treated by Cobalt-60 eye plaque. The median interval to cataract development is 4.2 years.
The radiation dose to the lens was also estimated from the computer-generated isodose curves. Using the original fundus drawings as a guide, the antero-posterior location of the tumor and the location of the lens were drawn on the isodose curve printout. This was done independently by two of the authors (M.K., J.A.). The dose and doserate to the lens were estimated for a point at the center of the lens. If calculations resulted in major discrepancies (i.e., differences of more than 20%), cases were reevaluated by two of the authors together (M.K., J.A.). The calculated dose to the lens ranged from 6.0 to 112.0 Gy (mean = 32.0 Gy, S.D. = 18.6 Gy) at a dose-rate ranging from 0.04 to 0.48 Gy/hr (mean = 0.15 Gy/hr, S.D. = 0.09 Gy/hr).
Analytical methods Actuarial estimates of the rate of cataract development over time were obtained by the product-limit method ( 13). The log rank test (16) was used to assess the statistical significance of a difference between the event rate curves of selected patient subgroups. Univariate and multivariate analysis of failure-time data were conducted using the Cox’s proportional hazards
The cumulative product-limit event rate curve for the development of cataract in 365 patients treated by Cobalt60 eye plaques is shown in Figure 1. The estimated probability of developing a cataract was 57% (SE. = 3%) at 5 years and 84% (S.E. = 3%) at 10 years. Table 2 shows the results of univariate analysis. The best univariate predictors of cataract development were thickness of the tumor and radiation dose to the lens, followed by dose-rate at the lens, largest basal tumor diameter, diameter of the plaque, location of the tumor’s anterior margin and duration of treatment. Age, location of the tumor’s posterior margin, and gender were all non significant (p 2 0.01) predictive factors by univariate analysis. To illustrate the univariate impact of radiation dose to the lens, we divided our patients into subgroups (Table 3) according to the tertiles of lens dose applied (cutpoints 20 Gy and 40 Gy), and then computed and plotted the 5-year post-plaque probabilities of developing a cataract (Fig. 2). The cumulative 5-year probability of developing a post-irradiation cataract was 0.22 for the low dose group (doses ranging from 6 to 20 Gy), 0.72 for the intermediate dose group (doses ranging from 20 to 40 Gy) and 0.81 for the high dose group (doses of 40 Gy or more). The differences between each of the three dose groups were highly significant (p < 0.0001, log rank test). The best combination of evaluated covariables for predicting cataractogenesis identified by multivariate Cox proportional hazard modeling included thickness of the tumor, the tumor’s anterior location and diameter of the plaque (Table 4). All treatment parameters except plaque
Table 2. Results of univariate prognostic factor analysis for the development of post-irradiation cataract in 365 patients with posterior uveal melanoma treated by Cobalt-60 plaque radiotherapy Variable
Beta
S.D.
t
Thickness of the tumor [mm] Dose to the lens [Gy] Dose-rate at the lens [Gy/hr] Largest basal tumor diameter [mm] Plaque’s diameter [ 10 mm vs. 15 mm] Anterior margin of the tumor [ 1, 2, 3 as described in Table l] Duration of treatment [hr] Posterior margin of the tumor [ 1, 2, 3 as described in Table l] Sex [males = 1, female = 21 Age [years1
0.32 0.0003 0.068 0.20 0.26 0.63 0.0062 0.32 0.32 0.0072
0.03 < 0.000 1 0.007 0.02 0.03 0.09 0.00 1 0.12 0.14 0.006
10.8 10.2 9.1 8.6 8.1 7.3 5.0 2.6 2.3 1.2
pvalue < < < < < < <
0.0001 0.0001 0.0001 0.000 1 0.000 1 0.000 1 0.0001 0.01 0.025 0.23
O'P
1000'0>
SE
sooo’o
1000’0
>
anp3A-d
S’S
1
I’0
PE’O
PO'0 PI'0 PO’0
oz.0
'a's wa8
[IWRL "IPaqF=JP se‘f'Z‘I]"O!lk?3OI 1oI.'a1uv [UIZU SI 'SA ulzuoI]anbeId30 lalauIe!a [““.“I ~a”2qq.L
wa!wd 59~aqlJop~!me wqa uo pawq s! ahIn yxg @Sal quw %o~‘~o~o > d) ~uwy!u%!s Qq@q s! ahIn yxa uaamlaq a3uaJagp aqL ‘Ldelaqo!par anbeld 09 -l@qoD %uunp suaI ayl 01 asop uogEgx3 (amour ~0 bf) m) y%!y JO (Lf, op 010~) awpauuaiu! ‘(Lg 0~ >) ~01 e %u+0aJ swag -ad u! ~~I?J~?~LQ %!dO[aAap 30 sa$tx @uenlx ahgt?Inurn~ ‘z ‘%!g
awpho3
%yapouI sp~~zeq pi?uo!vodoJd ,x03 dq payquap! [aporu a$tzutzAflInut Isag +QXIE alt?geAg[ncu 30 Ilnsax ‘p a[qeL
01
8
9
P
Z
0
w
ssu~puyaql 01 K.IoJ3~peJluo3 sysiqJ_*aKas'qlU! BUIOUE -lam B peq Kaq~leqlJ,~?J aq~jo al!dsu! ~uauwa~~jo pua .~a~jv s.nzaK s u~q~? J~VJEJWe padoIaAapsluayedaq~jo y.bZZ KIUO ‘popad yNM-Z 01 -1 e .nAO KIsnonuguoa pa.la -Agap6+-J ()z01 930 sasopuogI?IpE?.I JE ‘[email protected]~sIJdInS *saKa SB aq 01 .nzadde IOU saop Kq$@aq utpatunsse se jueuodw! ‘ssauy~~y~ SJ! put? lournl ayl jo u$h.xu ~ouawcz aqt 0% palelarro3 sry3p.j~ ‘sualaqlolasopaq~wa~ajyphraA aq 6w.usanbtqdaAgaX?o!pe.x L@M pawpeu! ewout?law leaAn %u~u~~uo3saKau~uo~~~n~~saq~aq~~sa%+nss~~nsa~m~ *saKaKq@?aq u! $seaIyt? ‘luapuadapasopK$3uo.1~s ST sysaua8o$at?.ngt?a JE~Ipaldaaat? Klpzaua%s! I!‘Kepol 3sop %uyzaJauyqlw ~uar.udolaAap 1312eie3jo aw aq3 u! awaz3urua ~0~s ola1qca.1a~11ys Kaq~‘sp.n2puoiss,Kt?pol -gnuI lsaqaql 01 papptzuaqM arqt?A aAga!paJdpa)y.uI Kq aAgtw!.Id al!nbaJaM sloqlneasaqlKq pasn spoqlatu jo s!suaIaq1 01 asop UO~E!~EJ(p)put?‘~a~atuerp anbyd -(LI) ~s61 UI ~vti~eue pup &au+sop aql *nowv put?uoy~~~~~own~'ssauy~~~~own~sapn~au~s~saua%o~ttl )qaod put?wt?pagq KqLrodal%uuaauo!d aq$ u! sualaql -ewa %ugznpaJd.IOJ salqt+n?A It?agnadwaq~o!pwpue @cy "Ojploy 01 UMOqS Is.ILJ SBM d!qsuoyqa~ Siq_L*anssyaq3 -ugaaugawqjo uogtzwquroa~saqaqt(a)‘wwuoa Kp~p u! SdOIaAaptt?qt a?huep aql ssalayl ‘anssgE 01 uoyt3p -Ial s! wauwa.r% la$jtz s.n?aK 01 ~s.~y aql%uunp alw $uaAa 43 jo asop aqi JaMol aq ‘ICpq1g3 pue K1pzapa~oay~ *q$oqlo‘aNna asuodsa1asop aql uo aql (q)‘q%q a$!nbsiluauIdoIaAap iae.nw?a jo aJt2.I @!.I%? alearpu! tztuouyauIun3.1%g asopp~oqsa~q~aq~papaa3xaaAoq@7~urs~uaw~t?a~~aq~jo -nw3 aAyqnwn3 aqt(t?)~oq~ -~wlo1.la~sodlojKd~~aqloyp~~anb~Idgg-yeqo3lay~s!sa -ua%owlwa uo KpnlsaAyaadso.IlaJ s!qijosynsaJaqL ~souuu! sasopaql~~ey~ wj aql ‘(sawL) [apow tsaqaql ut saIqeueAwwj@~s aqy put!saIqt?ueA asaqluaawaq uogtqa~oa %uo.~~saay~ o~anpaq@y aw-asopput! asop suaI.Ioj anp?AaA!la~pa.IdjOyat?~aq~ ySKpU&? a~t+n?Aynw mo uf ~~~e~ea uoyz~pwysod E jo $uawdoIaAap aq$ .IOJ sJowpald ~ut?ay@~sIOU alaM suaIaql 01 awl-asop pue asop uoyy3pt?l leqisn o~%u~s$.~nswqMawos SBM I! ‘snqJ'luauI~~a~laq)lalJesleaKo~ u~qlfM~acn2pauoye!p -EJe padolahapsuaIaqljo Jawa aql 1~ Kf) 0~ %u!paaa -xa asop uo~~tz~pt?l t!paA!aaaloqMsluayd 11e61lenp!A
Z'P
EO’O
-
I'Z
P8’0
0'1
*lapouI ayy.n?Agpu.u lsaqaq)01 pappE uaqM~olarpa-rd~uaayyu%!s $0~ aJaM a&wsop pue asop sual *uoywo~ J0u.w Jovawe putzssauya!ql10~1 -nlq)!Mpaqug uaqM anp3AaAg3ypaJdqaql Isollalawqp
18'0
SO'0
LS’O
EO’O
so-o
LS’O
80'0
0'6
SO’0
E6’0
co.0
L’E
swaged IIV (=o~~o Kf,OP)asOPrl%!H (43
ZL’O ZZ’O
op > 01 oz) asop aiepmaw (&OZ’ol9)~“PMO?
suaIaqloiasoa s*aL-0
s.reaA-s
1
:pal!syal 13aga alcJ-asop aq~ 'f 'a ‘Jauualg1.3 '1 ‘11~2~ ‘ZI 'ZL6IfOZZ-L8lw 53'30s'1% 'X 'IYUO!S -sn3s!p qly)saIqt?l ajgpuo s1apot.u uoyssa&Iax'a 'a ‘x03 ‘II 'ZS6I:OL-SSXPvwwdo 'WV 'V'yy'y'PEJ~JEDuo!le!pr?l jo ylssI.'alDEJeq3 @@oloqled pwp3Iug3.g ‘v ‘asaax :.a*a ‘uospIeuoat.3.a‘utioa '01 '6861f8Zl-LZI‘PZI ‘ZZIUO!sSnDSslp f[ZI--E[ I:E ‘TIIZdUOlS;rII!M &?OIO3UO 'BUIOU -eIaur Ieaanus spgl pue sluaurlea~l luaun3 'H .a ‘JBQ '6 '8061fLEI .ak?nvay3yy3suaursap
-6Z1:9p 3Il!ayua%nv'Iqm wM
jne ua~~ua%luao~llap%uny~!Mlnz'~ ‘pIajy3sIg-yI3.n~ '8 '06611889-Z89:IZ 'gJ"S '+u@ylqd~ wuoutqaur Kpoq hreg3.10Iep!o~oq~ hreurud "ojICdaraqlo!par anbyd 09-l@qo3snslaAuoyasal Ie+~ns jo Lpnlsdnos%paq3lag,tI '8 ‘uomuas!~f.3'[‘K.IM~-I :'f'a ‘laaIa*aw f.M *f‘Iaur\?f) f.x‘uazlywI 1.f'I‘la%mqs%nt/‘L '0661:19E-6SE‘6X-8PEX8 '3% 'wvwWo 'UW 'SU~Jl 'sdnotixapurcysou3oJdKq IeA!hlnsjo uosuedwo3 :ewou
~%wpuy Ino aln3aI
30 IuatudoIanap Igun dn-MOIIOJ30 q@ual pawwgsa ayl la%uoI ayj ‘IehIalu! dn-MO1103ayl la%uol ayl 5uoyt?nIeAa dn-MO1103ayJ30 %u!uy aql uo aaJ8ap auIOS01 luapuadap wM wz3ew paDnpuy-uogc!peJ e 30 uogDalap Igun auIP 30 q@uaI ayl ‘p.qq~ ‘lawa luaumall mo u! palerqelza -a.I a.IaMoyrn s$uagc?d ayl uo pasn aM ut?yl ~~?.n?lt?3uog -e;rpwysod %u~sou%t?~p “03 eua%pDwala3yp (asn p!p L(1qv -qold put?) pasn aAt?yh_u s)s$o~ouqwgydo %uwa3al ayL ‘@oIouqvylydo hrewud l!aql Icq dn-MOIIOJqayl30 wed w?aj w pey f&j’62 jeuoypp~ ue pue ‘Is~o~ouqey)qdo %u~a3aJ lraql tuog dn-MOIIOJqaql30 IIE pan!aDaJ pa!pnls aM s)uaytzd aql30 ohg’pj ‘KIsnOlAaJdpalw!pu! sv ‘sluap -ed 11~us Qu.u03!un apeur IOUstz~ Lpnls s!yt u! 19weit2~ uog~gw.~yod t?30 s~sou8mp ‘puoms *uogmu!isa asop 30 poqlam @3;[qdt?J% mo 30 .IO$uamayd anbeld 30 sa!semD mu! 01 anp ‘an@A IeaJ aqlao 1~3 aq ~y%!m sual ayl 01
(row0 > 4 99’0 (100.0 > 4 S9’0 Jalauwyps,anbeId
(100.0 > 4 ES’0 (100’0 > 4 IL’0 ssauy3qL
-I?Iatu IeaAnJogalsod103 uogeapnua snslaAMwaqlo!pw anbeIdi@qoD 'v 'f‘spIayS :'A .r‘Iau1"r3f.r.r‘la~~nqs%nt/‘9 ‘986 1f 199-SS9:POI ‘Io~@rl~rldo ‘WV *kPoq OFF’ Pue P!o.T -oqa aqljo sw.uoueIaur iueu%y~U 'ojddwaqlo!pw anbeId lIt?qo3 'SAuoylt?apnua'M '-I‘LpF3g:'v '[‘spIa!qs f'v"a %aquaaJg f.5'A ‘rplesf.M 'f ‘Iau.wf> f-1'1 ‘la%nqs%v ‘S '0661fZ6S-S8S:6OI 'Iouqeqlqdo'f'LUV TXuOUEIaUIItraAn loyalsodlojuogeapnua 'SAI(dwaqlo!peJ anbt?Ido9-lIeqo3 'M '~‘Icpe.~g f.~“uazlyw-~I.~'r‘Iau1er3f.r.r‘la%lnqs%nv ‘t '066If0091-96SI:99Ja3 -u~~%~Uou~Iau~lut?u@~U@aAn.~o~alsodql~Mslua~ledu! s~ol3~j cysou8oJdp?3!uq3'M 'f'IaUU?gt'f'r‘la%Jnqstinv ‘E '0661%1-6:1z%nS '+uwlqdo iy.10~luatulvaJ1 saoa :aorunllEIn30 'f 'r ‘la%lnqs%ny.Z '8861fl6P-P6VZl 'Iouqeqlqydg '1'~8'salter IeAy-tsjouosydruo3 :t?ruoueIatu Iaahn .10j uoyaI3nua SnsraAanbqd l@qo3 '-I‘s!ue~o~![ -0d ?a ‘l.~aqIv f.1‘uoppast's‘Jay3ed1.~ ‘p.~ojpaa ?CJ '8 ‘Y!t?H :'m 'a ‘q&lOMSII~ :'H 'a ‘UOSUWqV f's'x ‘SUIEpV .I
asop pawpDle3 aql s%uaywdawos “03 ‘lalzamoy faIq!ssod SE IC~a~mme St?suaj aql 01 awl-asop put2 asop aql alajn3Iw 01 svoJa aIq~ssod 1~ apm 3AEq aM ‘1s~~ *Aprilssgl30 sassau -yt?aM awos !no v+od 01 ayg pInoM pue azyu%o3al am ‘og-$lt?qo3 01 pa.IedtuoD ‘sual aq, st! qms ‘aLa paiEa.xlaq3 30 samwu~s ~emou ayl 01 asop uoy~!pe~ aq, ampal put! IeuosIad p+dsoq ar_p01 amsodxa uogt?!pw aqi az!ummu LayI :sanbtqd og-qeqo3 IaAo sa%iueApe p?ycmd &.& @cgaloaql qloq alzey 01 .wadde sanbqd g 1 -amp01 *s,o861 py ayl u! sanbtqd SZI-amp01 paplags 01 IClan!snjDxa~souq~ pay9liMs aM *asm ayl iou s! syy~ ywq Jo@al f! uo sanbyd og-$jr?qo3 sasn II!JSdnoH mo )eql amnsse @UI auo ‘saLa pa)e!ple.y anbyd og-y’t?qo3 I.#M s! JadEd syo us acwauadxa pavodal mo asnt?Daa -yDaur laJ)el aq1 ‘saba %ururewo+~otunl UI *(8 I) lap!auy3s . . -qoa bq 8261 u! passaJppe 92~ suogelalp2 ~e~~‘o~o~ -sliyd paanpu! uoy?!p~?.~pue a%urep .yn%t% ~~01103 JO suaI ayl 0% ak.ut?p lDa.up 01 hpuo3as s! u0yzeyp~1.1~ lal3E IuawdolaAap WLIWD .IaypyM 30 uogsanb ayl .suaI ay$30 neaI[m pm~o~o~s~yd aq$ ur sa%ut?q~aIq!ssod 01 %u!pk?ala% -ump anssg3o amyoh aq$ put! (la!awe!p anbyd) azrs play uo~~~~pw.~~ aq$ Aq pa)ouIoJd aq [email protected] suaI aq! 0) a%u.mp aql weyl %ugw!pu~ st? synsa1 mo la&aiu! pIno au0 -al r,@q LIangela~ e sagdtut pm ‘ayj~suaso~pel Llatua.x!xa aq 01 sua] aq$ %uysa%lns ‘amsodxa uop~!pw a@$ laye
(100.0 ’ 4 IL’0 (IO@0 ’ d) 69’0
suaI01 alw-asoa suaI 01 asoa
uog~30Ilogaluv
lalauxwps,anbeId aqlpue ‘ssauy3!ql ‘u$i.~s,row loualw sG~ot.unl aqlput2suaIaqljo Jalua3aqlol asop ‘alEJ-asop uaamlaq luapylao3uogeIa*lo3
‘s
aIqej_
630
13.
14.
15.
16.
I. J. Radiation Oncology 0 Biology0 Physics radiobiological considerations of importance in radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 2 1: 1403-1414;1991. Kaplan, E. L.; Meier, P. Nonparametric estimations from incomplete observations. J. Am. Stat. Assoc. 53:457481;1958. Lommatzsch, P. K.; Weise, B.; Ballin, R. Ein Beitrag zur Optimierung der Bestrahlungszeit bei der Behandlung des malignen Melanoms der Aderhaut mit beta-Applikatoren (106 Ru/106 Rh). Klin. Mbl. Augenheilk. 189: 133140;1986. Markoe, A. K.; Brady, L. W.; Karlsson, U. L.; Shields, .I. A.; Augsburger, J. J. Eye. In: Perez, C., Brady, L., eds. Principles and practice of radiation oncology. 2nd edition. Philadelphia, New York, London, Hagerstown: JB Lippincott; 1992. Mantel, N. Evaluation of survival data and two new rank order statistic arising in its consideration. Cancer Chemother. Rep. 50:163;1966.
Volume 26, Number 4, 1993 17. Merriam, G. R.; Focht, E. F. A clinical study of radiation cataracts and their relationship to dose. Am. J. Roentgenol. 77:759-785;1957. 18. Rohrschneider, W. Experimentelle Untersuchungen ueber die Veraenderungen normaler Augengewebe nach Roentgenbestrahlung. Graefe’s Archiv. Ophthalmol. 122:282298;1929. 19. Shields, J. A.; Shields, C. L.; Shah, P.; DePotter, P. Current alternatives in the management of posterior uveal melanomas. Trans. Pa. Acad. Ophthalmol. Otolaryngol. 42:938944; 1990. 20. Stallard, H. B. Radiotherapy for malignant melanoma of the choroid. Br. J. Ophthalmol. 50: 147- 155;1966. 2 1. Zimmerman, L. E.; McLean, I. W.; Forster, W. D. Does enucleation of the eye containing a malignant melanoma prevent or accelerate the dissemination of tumor cells? Br. J. Ophthalmol. 62:420-425; 1978.
Inr. J. Radiation Oncology Bml. Phys., Vol. 26, pp. 625-630 Printed in the U.S.A. All rights reserved.
??Clinical Original Contribution
CATARACTOGENESIS MARKUS
KLEINEIDAM,
AFTER COBALT-60 EYE PLAQUE RADIOTHERAPY
M.D.,’ JAMES J. AUGSBURGER, M.D.,’ CARLOS HERNANDEZ, M.D.,2
PATRICK GLENNON, M.S.2 AND LUTHER W. BRADY, M.D.2 ‘Oncology Unit, Retina Service, Wills Eye Hospital, Jefferson Medical College; and %he Department of Radiation Oncology and Nuclear Medicine, Hahnemann University, Philadelphia, Pennsylvania, USA Purpose: This studywas designed to estimate the actuarial incidence of typical postirradiation cataracts and to identifyprognostic factors related to their development in melanoma-containing eyes treated by Cobalt-60 plaque radiotherapy. Our special interest was the impact of calculated radiation dose and dose-rate to the lens. MethodsandMaterial:The authorsevaluated theactuarial occurrence of post-irradiation cataract in 365 patients with primary posterior uveal melanoma treated by Cobalt-60 plaque radiotherapy between 1976 and 1986. Results: Only 22% (S.E. = 4.6%) of the patients who received a total dose of 6 to 20 Gy at the center of the lens developed a visually significant cataract attributable to the radiation within 5 years after treatment. Using multivariate Cox proportional hazards modeling, the authors identified thickness of the tumor, location of the tumor’s anterior margin relative to the equator and the ora serrata, and diameter of the eye plaque used as the best combination of covariables for predicting length of time until development of cataract. Surprisingly, the dose of radiation delivered to the lens, which was strongly correlated to all of these covariables, was not a significant predictive factor in multivariate analysis. Conclusion: The results suggest that success of efforts to decrease the occurrence rate of post-irradiation cataracts by better treatment planning might be limited in patients with posterior uveal melanoma. Uveal melanoma, Cataract, Brachytherapy,
Cobalt-60 eye plaques.
INTRODUCTION
1952 (lo), include posterior subcapsular lens clouding caused by damage to the subcapsular epithelium. In spite of the fact that post-irradiation cataracts are well-known to ophthalmologists, the actuarial incidence of this form of cataract following specific methods of ocular radiotherapy has not been reported. This study was designed to estimate the actuarial incidence of typical postirradiation cataracts and to identify prognostic factors related to their development in melanoma-containing eyes treated by Cobalt-60 plaque radiotherapy. Our special interest was the impact of calculated radiation dose and dose-rate (12) to the lens.
On the basis of currently available, non randomized but statistically-adjusted comparative survival studies, episcleral plaque radiotherapy and enucleation (21) are believed to be nearly equivalent in terms of the survival prognosis of patients having a posterior uveal malignant melanoma ( 1-7, 9, 19). If this is true, plaque treatment has the advantage of preserving vision, at least to some degree, in the tumor-containing eye in a substantial proportion of patients for intervals of up to 5 years or even longer. Unfortunately, the great majority of patients treated by episcleral plaque radiotherapy lose a substantial portion of their vision over 5- 10 years, largely on the basis of radiation-related complications ( 15). One of the most widely recognized and common complications of episcleral plaque brachytherapy in eyes with primary posterior uveal melanoma is cataract formation (8, 14, 15, 17). The classic features of post-irradiation cataract, as described by Cogan, Donaldson and Reese in
Four hundred patients with posterior uveal malignant melanoma were treated primarily by Cobalt-60 plaque radiotherapy at Wills Eye Hospital, Philadelphia, between
Reprint requests to: James J. Augsburger, M.D., Oncology Unit, Retina Service, Wills Eye Hospital, 900 Walnut St., Philadelphia, PA 19 107, USA. Acknowledgement: The authors would like to acknowledge Jerry A. Shields, M.D., who managed some of the patients evaluated in this study.
This study was supported in part by a grant from the Deutsche Krebsgesellschaft, Dr. Mildred Scheel Stiftung, Bonn, Germany, to Dr. Markus Kleineidam, and an unrestricted grant from Research to Prevent Blindness, Inc., New York, NY, U.S.A., to Wills Eye Hospital. Accepted for publication 15 February 1993.
METHODS
AND MATERIALS
Patients
625
1. J. Radiation Oncology 0 Biology 0 Physics
626
1976 and 1986. Eighteen patients received additional plaque radiotherapy for local tumor relapse subsequent to the original radiotherapy and were therefore excluded from this study. The remaining 382 patients (Table 1) included 186 women (48.7%) and 196 (51.3%) men. These patients ranged from 25 to 89 years of age and had a mean age of 59 years. Tumor thickness ranged from 1.5 to 15.5 mm (mean = 6.5 mm, S.D. = 2.4 mm) and largest basal tumor diameter ranged from 5.0 to 22.0 mm (mean = 11.5 mm, S.D. = 3.0 mm). The anterior margin of the tumor was posterior to the equator in 145 eyes, between the equator and ora serrata in 139 eyes and anterior to the ora serrata in 98 eyes. The posterior tumor margin was located within 3 mm of the optic disc in 113 eyes, over 3 mm from the optic disc but posterior to the ocular equator in 246 eyes, and anterior to the equator in 23 eyes. Seventeen of the 382 patients (4.5%) had undergone cataract surgery prior to their plaque radiotherapy and therefore were not at risk to develop a post-irradiation cataract. These 17 patients were excluded prior to final data analysis, leaving us with a study group of 365 patients. Follow-up and endpoint
Patients were followed up until death or their most recent re-examination before the closing date for data analysis (May 1992). Some patients who were unable to return to Wills Eye Hospital for regular periodic followup examination were advised to undergo periodic re-examination and evaluation by their referring ophthalmologist. Two hundred two of the 365 patients (55.3%) had all of their follow-up at Wills Eye Hospital. In contrast, 109 (39.9%) had some of their follow-up at Wills and the rest with their referring ophthalmologist (usually on an
Volume 26, Number 4, I993 visit basis), and 54 (14.8%) had all of their follow-up performed by the referring ophthalmologist. Patients were considered to be lost to follow-up if they had not been re-examined at Wills Eye Hospital or by their local ophthalmologist within 1 year prior to the closing date. For the purposes of this study, we defined a post-irradiation cataract as lens clouding subsequent to radiotherapy which (a) included posterior subcapsular changes as a prominent feature (lo), (b) was of sufficient extent to be a contributing factor to post-irradiation visual impairment, and (c) was asymmetrically more pronounced than any concurrent lens clouding in the fellow eye. No staging of lens clouding was attempted. The endpoint assessed in this study was length of post treatment followup until diagnosis of a post-irradiation cataract.
alternating
Cobalt-60 eye plaque radiotherapy
The Cobalt-60 applicators used were developed by Stallard in 1966 (20). Only plaques of 10 mm (160 cases) and 15 mm (222 cases) were used in this series. Further details have been published previously (5, 15). The radiation doses to the tumor’s apex and base and the overall treatment time were estimated from two-dimensional isodose distributions generated by a computer program (PRIME). The estimated dose-rate 1 mm from the surface of the plaque (base of the tumor) ranged from 0.88 to 2.61 Gy/hr (mean = I .72 Gy/hr, S.D. = 0.27 Gy/ hr). The overall treatment time ranged from 81 to 432 hrs (mean = 211 hrs, S.D. = 55.0 hrs). The basal radiation dose ranged from 171.0 Gy to 702.0 Gy (mean = 355.0 Gy, S.D. = 76.0 Gy) and the apical radiation dose ranged from 26.3 Gy to 157.0 Gy (mean = 83.5 Gy, S.D. = 17.6 Gy).
Table 1. Variables entered in the analysis and their distributions Variable
Mean
S.D.
Range
Age Dose to the Lens Dose-rate at the Lens Duration of Treatment Largest basal Tumor Diameter Thickness of the Tumor
59 years 32.0 Gy 0.15 Gy/h 211 h 11.5 mm 6.5 mm
12.7 18.6 0.09 55.0 3.0 2.4
25-89 years 6-112 Gy 0.04-0.48 Gy/h 81-432 h 5-22 mm 1.5-15.5 mm
Anterior location
1. Posterior to the equator 2. Posterior to the ora serrata 3. Anterior to the ora serrata 1. 3 mm from disc but posterior to the equator 3. Anterior to the equator 10 mm: 160 (41.88%) 15 mm: 222 (58.12%) 196 (51.31%) Male Female 186 (48.69%)
Posterior location
Plaque’s diameter Sex
145 (37.96%) 139 (36.39%) 98 (25.65%) 113 (29.58%) 246 (64.40%) 23 (6.02%)
621
Cataractogenesis after Co-60 plaque radiotherapy 0 M. KLEINEIDAM et al.
;
0.6
model (11). In the multivariate analysis stepwise stepdown regression was carried out, starting with all covariables and excluding one at a time those not associated with a p-value of at least 0.0 1. This final model was then tested against other models incorporating covariables
B $ d
0.4
which were of high predictive analysis.
8 b
0.2
s
0.6
9
E 0 8 !? P
0.0 0
value
in the univariate
RESULTS
2
4
6
6
10
Time after treatment [years]
Fig. 1. Cumulative actuarial rates of developing post-irradiation cataract in 365 patients with primary posterior uveal melanoma treated by Cobalt-60 eye plaque. The median interval to cataract development is 4.2 years.
The radiation dose to the lens was also estimated from the computer-generated isodose curves. Using the original fundus drawings as a guide, the antero-posterior location of the tumor and the location of the lens were drawn on the isodose curve printout. This was done independently by two of the authors (M.K., J.A.). The dose and doserate to the lens were estimated for a point at the center of the lens. If calculations resulted in major discrepancies (i.e., differences of more than 20%), cases were reevaluated by two of the authors together (M.K., J.A.). The calculated dose to the lens ranged from 6.0 to 112.0 Gy (mean = 32.0 Gy, S.D. = 18.6 Gy) at a dose-rate ranging from 0.04 to 0.48 Gy/hr (mean = 0.15 Gy/hr, S.D. = 0.09 Gy/hr).
Analytical methods Actuarial estimates of the rate of cataract development over time were obtained by the product-limit method ( 13). The log rank test (16) was used to assess the statistical significance of a difference between the event rate curves of selected patient subgroups. Univariate and multivariate analysis of failure-time data were conducted using the Cox’s proportional hazards
The cumulative product-limit event rate curve for the development of cataract in 365 patients treated by Cobalt60 eye plaques is shown in Figure 1. The estimated probability of developing a cataract was 57% (SE. = 3%) at 5 years and 84% (S.E. = 3%) at 10 years. Table 2 shows the results of univariate analysis. The best univariate predictors of cataract development were thickness of the tumor and radiation dose to the lens, followed by dose-rate at the lens, largest basal tumor diameter, diameter of the plaque, location of the tumor’s anterior margin and duration of treatment. Age, location of the tumor’s posterior margin, and gender were all non significant (p 2 0.01) predictive factors by univariate analysis. To illustrate the univariate impact of radiation dose to the lens, we divided our patients into subgroups (Table 3) according to the tertiles of lens dose applied (cutpoints 20 Gy and 40 Gy), and then computed and plotted the 5-year post-plaque probabilities of developing a cataract (Fig. 2). The cumulative 5-year probability of developing a post-irradiation cataract was 0.22 for the low dose group (doses ranging from 6 to 20 Gy), 0.72 for the intermediate dose group (doses ranging from 20 to 40 Gy) and 0.81 for the high dose group (doses of 40 Gy or more). The differences between each of the three dose groups were highly significant (p < 0.0001, log rank test). The best combination of evaluated covariables for predicting cataractogenesis identified by multivariate Cox proportional hazard modeling included thickness of the tumor, the tumor’s anterior location and diameter of the plaque (Table 4). All treatment parameters except plaque
Table 2. Results of univariate prognostic factor analysis for the development of post-irradiation cataract in 365 patients with posterior uveal melanoma treated by Cobalt-60 plaque radiotherapy Variable
Beta
S.D.
t
Thickness of the tumor [mm] Dose to the lens [Gy] Dose-rate at the lens [Gy/hr] Largest basal tumor diameter [mm] Plaque’s diameter [ 10 mm vs. 15 mm] Anterior margin of the tumor [ 1, 2, 3 as described in Table l] Duration of treatment [hr] Posterior margin of the tumor [ 1, 2, 3 as described in Table l] Sex [males = 1, female = 21 Age [years1
0.32 0.0003 0.068 0.20 0.26 0.63 0.0062 0.32 0.32 0.0072
0.03 < 0.000 1 0.007 0.02 0.03 0.09 0.00 1 0.12 0.14 0.006
10.8 10.2 9.1 8.6 8.1 7.3 5.0 2.6 2.3 1.2
pvalue < < < < < < <
0.0001 0.0001 0.0001 0.000 1 0.000 1 0.000 1 0.0001 0.01 0.025 0.23
O'P
1000'0>
SE
sooo’o
1000’0
>
anp3A-d
S’S
1
I’0
PE’O
PO'0 PI'0 PO’0
oz.0
'a's wa8
[IWRL "IPaqF=JP se‘f'Z‘I]"O!lk?3OI 1oI.'a1uv [UIZU SI 'SA ulzuoI]anbeId30 lalauIe!a [““.“I ~a”2qq.L
wa!wd 59~aqlJop~!me wqa uo pawq s! ahIn yxg @Sal quw %o~‘~o~o > d) ~uwy!u%!s Qq@q s! ahIn yxa uaamlaq a3uaJagp aqL ‘Ldelaqo!par anbeld 09 -l@qoD %uunp suaI ayl 01 asop uogEgx3 (amour ~0 bf) m) y%!y JO (Lf, op 010~) awpauuaiu! ‘(Lg 0~ >) ~01 e %u+0aJ swag -ad u! ~~I?J~?~LQ %!dO[aAap 30 sa$tx @uenlx ahgt?Inurn~ ‘z ‘%!g
awpho3
%yapouI sp~~zeq pi?uo!vodoJd ,x03 dq payquap! [aporu a$tzutzAflInut Isag +QXIE alt?geAg[ncu 30 Ilnsax ‘p a[qeL
01
8
9
P
Z
0
w
ssu~puyaql 01 K.IoJ3~peJluo3 sysiqJ_*aKas'qlU! BUIOUE -lam B peq Kaq~leqlJ,~?J aq~jo al!dsu! ~uauwa~~jo pua .~a~jv s.nzaK s u~q~? J~VJEJWe padoIaAapsluayedaq~jo y.bZZ KIUO ‘popad yNM-Z 01 -1 e .nAO KIsnonuguoa pa.la -Agap6+-J ()z01 930 sasopuogI?IpE?.I JE ‘[email protected]~sIJdInS *saKa SB aq 01 .nzadde IOU saop Kq$@aq utpatunsse se jueuodw! ‘ssauy~~y~ SJ! put? lournl ayl jo u$h.xu ~ouawcz aqt 0% palelarro3 sry3p.j~ ‘sualaqlolasopaq~wa~ajyphraA aq 6w.usanbtqdaAgaX?o!pe.x L@M pawpeu! ewout?law leaAn %u~u~~uo3saKau~uo~~~n~~saq~aq~~sa%+nss~~nsa~m~ *saKaKq@?aq u! $seaIyt? ‘luapuadapasopK$3uo.1~s ST sysaua8o$at?.ngt?a JE~Ipaldaaat? Klpzaua%s! I!‘Kepol 3sop %uyzaJauyqlw ~uar.udolaAap 1312eie3jo aw aq3 u! awaz3urua ~0~s ola1qca.1a~11ys Kaq~‘sp.n2puoiss,Kt?pol -gnuI lsaqaql 01 papptzuaqM arqt?A aAga!paJdpa)y.uI Kq aAgtw!.Id al!nbaJaM sloqlneasaqlKq pasn spoqlatu jo s!suaIaq1 01 asop UO~E!~EJ(p)put?‘~a~atuerp anbyd -(LI) ~s61 UI ~vti~eue pup &au+sop aql *nowv put?uoy~~~~~own~'ssauy~~~~own~sapn~au~s~saua%o~ttl )qaod put?wt?pagq KqLrodal%uuaauo!d aq$ u! sualaql -ewa %ugznpaJd.IOJ salqt+n?A It?agnadwaq~o!pwpue @cy "Ojploy 01 UMOqS Is.ILJ SBM d!qsuoyqa~ Siq_L*anssyaq3 -ugaaugawqjo uogtzwquroa~saqaqt(a)‘wwuoa Kp~p u! SdOIaAaptt?qt a?huep aql ssalayl ‘anssgE 01 uoyt3p -Ial s! wauwa.r% la$jtz s.n?aK 01 ~s.~y aql%uunp alw $uaAa 43 jo asop aqi JaMol aq ‘ICpq1g3 pue K1pzapa~oay~ *q$oqlo‘aNna asuodsa1asop aql uo aql (q)‘q%q a$!nbsiluauIdoIaAap iae.nw?a jo aJt2.I @!.I%? alearpu! tztuouyauIun3.1%g asopp~oqsa~q~aq~papaa3xaaAoq@7~urs~uaw~t?a~~aq~jo -nw3 aAyqnwn3 aqt(t?)~oq~ -~wlo1.la~sodlojKd~~aqloyp~~anb~Idgg-yeqo3lay~s!sa -ua%owlwa uo KpnlsaAyaadso.IlaJ s!qijosynsaJaqL ~souuu! sasopaql~~ey~ wj aql ‘(sawL) [apow tsaqaql ut saIqeueAwwj@~s aqy put!saIqt?ueA asaqluaawaq uogtqa~oa %uo.~~saay~ o~anpaq@y aw-asopput! asop suaI.Ioj anp?AaA!la~pa.IdjOyat?~aq~ ySKpU&? a~t+n?Aynw mo uf ~~~e~ea uoyz~pwysod E jo $uawdoIaAap aq$ .IOJ sJowpald ~ut?ay@~sIOU alaM suaIaql 01 awl-asop pue asop uoyy3pt?l leqisn o~%u~s$.~nswqMawos SBM I! ‘snqJ'luauI~~a~laq)lalJesleaKo~ u~qlfM~acn2pauoye!p -EJe padolahapsuaIaqljo Jawa aql 1~ Kf) 0~ %u!paaa -xa asop uo~~tz~pt?l t!paA!aaaloqMsluayd 11e61lenp!A
Z'P
EO’O
-
I'Z
P8’0
0'1
*lapouI ayy.n?Agpu.u lsaqaq)01 pappE uaqM~olarpa-rd~uaayyu%!s $0~ aJaM a&wsop pue asop sual *uoywo~ J0u.w Jovawe putzssauya!ql10~1 -nlq)!Mpaqug uaqM anp3AaAg3ypaJdqaql Isollalawqp
18'0
SO'0
LS’O
EO’O
so-o
LS’O
80'0
0'6
SO’0
E6’0
co.0
L’E
swaged IIV (=o~~o Kf,OP)asOPrl%!H (43
ZL’O ZZ’O
op > 01 oz) asop aiepmaw (&OZ’ol9)~“PMO?
suaIaqloiasoa s*aL-0
s.reaA-s
1
:pal!syal 13aga alcJ-asop aq~ 'f 'a ‘Jauualg1.3 '1 ‘11~2~ ‘ZI 'ZL6IfOZZ-L8lw 53'30s'1% 'X 'IYUO!S -sn3s!p qly)saIqt?l ajgpuo s1apot.u uoyssa&Iax'a 'a ‘x03 ‘II 'ZS6I:OL-SSXPvwwdo 'WV 'V'yy'y'PEJ~JEDuo!le!pr?l jo ylssI.'alDEJeq3 @@oloqled pwp3Iug3.g ‘v ‘asaax :.a*a ‘uospIeuoat.3.a‘utioa '01 '6861f8Zl-LZI‘PZI ‘ZZIUO!sSnDSslp f[ZI--E[ I:E ‘TIIZdUOlS;rII!M &?OIO3UO 'BUIOU -eIaur Ieaanus spgl pue sluaurlea~l luaun3 'H .a ‘JBQ '6 '8061fLEI .ak?nvay3yy3suaursap
-6Z1:9p 3Il!ayua%nv'Iqm wM
jne ua~~ua%luao~llap%uny~!Mlnz'~ ‘pIajy3sIg-yI3.n~ '8 '06611889-Z89:IZ 'gJ"S '+u@ylqd~ wuoutqaur Kpoq hreg3.10Iep!o~oq~ hreurud "ojICdaraqlo!par anbyd 09-l@qo3snslaAuoyasal Ie+~ns jo Lpnlsdnos%paq3lag,tI '8 ‘uomuas!~f.3'[‘K.IM~-I :'f'a ‘laaIa*aw f.M *f‘Iaur\?f) f.x‘uazlywI 1.f'I‘la%mqs%nt/‘L '0661:19E-6SE‘6X-8PEX8 '3% 'wvwWo 'UW 'SU~Jl 'sdnotixapurcysou3oJdKq IeA!hlnsjo uosuedwo3 :ewou
~%wpuy Ino aln3aI
30 IuatudoIanap Igun dn-MOIIOJ30 q@ual pawwgsa ayl la%uoI ayj ‘IehIalu! dn-MO1103ayl la%uol ayl 5uoyt?nIeAa dn-MO1103ayJ30 %u!uy aql uo aaJ8ap auIOS01 luapuadap wM wz3ew paDnpuy-uogc!peJ e 30 uogDalap Igun auIP 30 q@uaI ayl ‘p.qq~ ‘lawa luaumall mo u! palerqelza -a.I a.IaMoyrn s$uagc?d ayl uo pasn aM ut?yl ~~?.n?lt?3uog -e;rpwysod %u~sou%t?~p “03 eua%pDwala3yp (asn p!p L(1qv -qold put?) pasn aAt?yh_u s)s$o~ouqwgydo %uwa3al ayL ‘@oIouqvylydo hrewud l!aql Icq dn-MOIIOJqayl30 wed w?aj w pey f&j’62 jeuoypp~ ue pue ‘Is~o~ouqey)qdo %u~a3aJ lraql tuog dn-MOIIOJqaql30 IIE pan!aDaJ pa!pnls aM s)uaytzd aql30 ohg’pj ‘KIsnOlAaJdpalw!pu! sv ‘sluap -ed 11~us Qu.u03!un apeur IOUstz~ Lpnls s!yt u! 19weit2~ uog~gw.~yod t?30 s~sou8mp ‘puoms *uogmu!isa asop 30 poqlam @3;[qdt?J% mo 30 .IO$uamayd anbeld 30 sa!semD mu! 01 anp ‘an@A IeaJ aqlao 1~3 aq ~y%!m sual ayl 01
(row0 > 4 99’0 (100.0 > 4 S9’0 Jalauwyps,anbeId
(100.0 > 4 ES’0 (100’0 > 4 IL’0 ssauy3qL
-I?Iatu IeaAnJogalsod103 uogeapnua snslaAMwaqlo!pw anbeIdi@qoD 'v 'f‘spIayS :'A .r‘Iau1"r3f.r.r‘la~~nqs%nt/‘9 ‘986 1f 199-SS9:POI ‘Io~@rl~rldo ‘WV *kPoq OFF’ Pue P!o.T -oqa aqljo sw.uoueIaur iueu%y~U 'ojddwaqlo!pw anbeId lIt?qo3 'SAuoylt?apnua'M '-I‘LpF3g:'v '[‘spIa!qs f'v"a %aquaaJg f.5'A ‘rplesf.M 'f ‘Iau.wf> f-1'1 ‘la%nqs%v ‘S '0661fZ6S-S8S:6OI 'Iouqeqlqdo'f'LUV TXuOUEIaUIItraAn loyalsodlojuogeapnua 'SAI(dwaqlo!peJ anbt?Ido9-lIeqo3 'M '~‘Icpe.~g f.~“uazlyw-~I.~'r‘Iau1er3f.r.r‘la%lnqs%nv ‘t '066If0091-96SI:99Ja3 -u~~%~Uou~Iau~lut?u@~U@aAn.~o~alsodql~Mslua~ledu! s~ol3~j cysou8oJdp?3!uq3'M 'f'IaUU?gt'f'r‘la%Jnqstinv ‘E '0661%1-6:1z%nS '+uwlqdo iy.10~luatulvaJ1 saoa :aorunllEIn30 'f 'r ‘la%lnqs%ny.Z '8861fl6P-P6VZl 'Iouqeqlqydg '1'~8'salter IeAy-tsjouosydruo3 :t?ruoueIatu Iaahn .10j uoyaI3nua SnsraAanbqd l@qo3 '-I‘s!ue~o~![ -0d ?a ‘l.~aqIv f.1‘uoppast's‘Jay3ed1.~ ‘p.~ojpaa ?CJ '8 ‘Y!t?H :'m 'a ‘q&lOMSII~ :'H 'a ‘UOSUWqV f's'x ‘SUIEpV .I
asop pawpDle3 aql s%uaywdawos “03 ‘lalzamoy faIq!ssod SE IC~a~mme St?suaj aql 01 awl-asop put2 asop aql alajn3Iw 01 svoJa aIq~ssod 1~ apm 3AEq aM ‘1s~~ *Aprilssgl30 sassau -yt?aM awos !no v+od 01 ayg pInoM pue azyu%o3al am ‘og-$lt?qo3 01 pa.IedtuoD ‘sual aq, st! qms ‘aLa paiEa.xlaq3 30 samwu~s ~emou ayl 01 asop uoy~!pe~ aq, ampal put! IeuosIad p+dsoq ar_p01 amsodxa uogt?!pw aqi az!ummu LayI :sanbtqd og-qeqo3 IaAo sa%iueApe p?ycmd &.& @cgaloaql qloq alzey 01 .wadde sanbqd g 1 -amp01 *s,o861 py ayl u! sanbtqd SZI-amp01 paplags 01 IClan!snjDxa~souq~ pay9liMs aM *asm ayl iou s! syy~ ywq Jo@al f! uo sanbyd og-$jr?qo3 sasn II!JSdnoH mo )eql amnsse @UI auo ‘saLa pa)e!ple.y anbyd og-y’t?qo3 I.#M s! JadEd syo us acwauadxa pavodal mo asnt?Daa -yDaur laJ)el aq1 ‘saba %ururewo+~otunl UI *(8 I) lap!auy3s . . -qoa bq 8261 u! passaJppe 92~ suogelalp2 ~e~~‘o~o~ -sliyd paanpu! uoy?!p~?.~pue a%urep .yn%t% ~~01103 JO suaI ayl 0% ak.ut?p lDa.up 01 hpuo3as s! u0yzeyp~1.1~ lal3E IuawdolaAap WLIWD .IaypyM 30 uogsanb ayl .suaI ay$30 neaI[m pm~o~o~s~yd aq$ ur sa%ut?q~aIq!ssod 01 %u!pk?ala% -ump anssg3o amyoh aq$ put! (la!awe!p anbyd) azrs play uo~~~~pw.~~ aq$ Aq pa)ouIoJd aq [email protected] suaI aq! 0) a%u.mp aql weyl %ugw!pu~ st? synsa1 mo la&aiu! pIno au0 -al r,@q LIangela~ e sagdtut pm ‘ayj~suaso~pel Llatua.x!xa aq 01 sua] aq$ %uysa%lns ‘amsodxa uop~!pw a@$ laye
(100.0 ’ 4 IL’0 (IO@0 ’ d) 69’0
suaI01 alw-asoa suaI 01 asoa
uog~30Ilogaluv
lalauxwps,anbeId aqlpue ‘ssauy3!ql ‘u$i.~s,row loualw sG~ot.unl aqlput2suaIaqljo Jalua3aqlol asop ‘alEJ-asop uaamlaq luapylao3uogeIa*lo3
‘s
aIqej_
630
13.
14.
15.
16.
I. J. Radiation Oncology 0 Biology0 Physics radiobiological considerations of importance in radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 2 1: 1403-1414;1991. Kaplan, E. L.; Meier, P. Nonparametric estimations from incomplete observations. J. Am. Stat. Assoc. 53:457481;1958. Lommatzsch, P. K.; Weise, B.; Ballin, R. Ein Beitrag zur Optimierung der Bestrahlungszeit bei der Behandlung des malignen Melanoms der Aderhaut mit beta-Applikatoren (106 Ru/106 Rh). Klin. Mbl. Augenheilk. 189: 133140;1986. Markoe, A. K.; Brady, L. W.; Karlsson, U. L.; Shields, .I. A.; Augsburger, J. J. Eye. In: Perez, C., Brady, L., eds. Principles and practice of radiation oncology. 2nd edition. Philadelphia, New York, London, Hagerstown: JB Lippincott; 1992. Mantel, N. Evaluation of survival data and two new rank order statistic arising in its consideration. Cancer Chemother. Rep. 50:163;1966.
Volume 26, Number 4, 1993 17. Merriam, G. R.; Focht, E. F. A clinical study of radiation cataracts and their relationship to dose. Am. J. Roentgenol. 77:759-785;1957. 18. Rohrschneider, W. Experimentelle Untersuchungen ueber die Veraenderungen normaler Augengewebe nach Roentgenbestrahlung. Graefe’s Archiv. Ophthalmol. 122:282298;1929. 19. Shields, J. A.; Shields, C. L.; Shah, P.; DePotter, P. Current alternatives in the management of posterior uveal melanomas. Trans. Pa. Acad. Ophthalmol. Otolaryngol. 42:938944; 1990. 20. Stallard, H. B. Radiotherapy for malignant melanoma of the choroid. Br. J. Ophthalmol. 50: 147- 155;1966. 2 1. Zimmerman, L. E.; McLean, I. W.; Forster, W. D. Does enucleation of the eye containing a malignant melanoma prevent or accelerate the dissemination of tumor cells? Br. J. Ophthalmol. 62:420-425; 1978.