~ 1992 ~lse~cr Sc~e~c~ publishers ~V. All ~ights rcse~cd. 0 I67-814g'92,'S05 00
223
RADION 00957
Iridium-192 brachytherapy in the management of 147 T2N o oral tongue carcinomas treated with irradiation alone: comparison of two
treatment techniques M . p e r n o t , L . M a l i s s a r d , P . Aletti, S. Hoffstetter, J . J . F o r w a r d a n d P. B e y (ReceJ~d 19 Aprlt 1991, re~sion received 10 October 1991, ae~pt ed 5 D~ember 1991)
Summary Our purpose is to analyse local control, comphcatlons relative to the proportion of ~otal dose delivered by external beam i~adiationvcrsusintersti~alimphntinl47pat~e~sw~hprev~o~s~]~ntrcatedT~N~sq~am~ns~e1l~arein~ma~fthe~ralt~ag~e~
managed b~ween 1973 and 1986 (UICC ~ta~Jng system) These T~ NOpatients are part of a larger group of 430 p a l l e t s with o ral tongue carcinoma (T,, T:, T~ ) treated with irradiation ~I~e. Of these 1~7 T 2 N Op atients, 70 wQre trc~ted with interstiUal implant alone and 77 wkh both externaI beam irradiation a n d ~mptant. In th* group treated with interstitial i m p l ~ t alone, the 5-year Iocal control was 89.8 Z against 50.6 ~o in t h e e treated with extemz] beam irradiation and inter~titlal implant (log-rank test, p - 0.00002); 67.6~ versus 46.5% for Ioeoreglonal control (p = O.029); and 62.2Z vergus 34.7 po for specific survival ( p - 0.0015), S~nce 1980, nit the p~tients treated by iridium implantation were protected w k h a I ~ d e d spacing devise between t he tongue and the m an dible. Soft tlssue n~rosls'and bone r xposu re following t ~ t m c n t ~ = scored ae~rdlng 1o the foLLo~ng criteria: minor, moderate or severe. Seven moderate and one severe ~mpLieati~s were ~eorded in the brachytherapy ~roap. None oC~h= patients requi~d surgery. In the combined tr~atm~t g~oup, six moderate and two s~ere complications w ~ e observed. Patlen ts treate d with interstitial i m p l ~ t alone, ~ d showing moderate or severe complications h ad received ~ average braehytherapy dose of 7600cGy. In the s ~ e group, the patients ~ t h o n t ~mplications had received ~ average dose of 6800 cGy. In the combined t ~atment group, the averag~ e~emal radiation dos9 and the awrage braeh~h0rapy d o ~ w ~ not significantly different whether or not the patients had complications, but we noted an improvement in local control when the major part of the dose was delivered by interstitla[ impLant. We cur~ntly recommend brachy~h~rapy ~]one for the initial treatment of T~ N,, cancers of the oral tongge, but an elective functional neck dissection is essential for these patients to a ~ i d subsequent neck node metastases.
ln~,oduetion C a r c i n o m a of t i c mobile tongue c a n b e treated b y s ~ gery m d radiotherapy, b y external radiotherapy alone, b y externaL radiotherapy a n d brachytberapy, or b y b r ~ h y t h ~ p y alone, with or without n e c k dissection. A n y role for a d j u v a n t c h e m o t h ~ a p y is n o t yct proven, I n 1972, w e a b ~ d o n e d radium, in favor o f iridium a n d u s u a g y t r e a t e d oral tongue earalnoma without alinicat a d e n o p a t h y b y b r a c h y t h ~ a p y alone, with or without n ~ k dissection. Patients with nodal involvement were Addressfor co~po;,den~: Dr, M. Pernot. Sr
treated b y e• b e a m irradiation with a brachytherapy boost to the primary site, with n e c k diss~:tion, if n ~ e s s a r y , to address residual ceralca] disease. Surgery w a s ~ s e ~ e d for very [ ~ g e tumors n e ~ the m ~ d i b[e or with several mobile nodes. During 1975, thflucnced b y C h u ~ d F l ~ t c h ~ [21, %o began utilizing external irradiation prior to brachytherapy even for T t No a n d T 2 No presentations. W e soon noticed ~ i n c ~ a s e in ~currences at the pdmmry site ~ d ~sume~i t r c a t m ~ t of'these two disease stages with braehytherapy alone for the primary tumor. Some
de Cufi~h~rapie, Centre Alexis Vautfin. 545 [ t Vand~u~e-I~s-Na.oy Cede, Fr~ce.
224
radiotherapists in our institution, who were not convinced by these ialtla/results, have continued the assoal~don of external irradiation and braehythempy, even for T 1TmNn 0~sentabons. Therefore, we have two parallel ser~es of Tz No patients, but ~thout randomi* zadon. The aim of this ~vork is to analyse the Iocoreglonal control and sufalval of these two group~ and to att erupt an exgianzdon of reeu~enee patterns and eompllcations Materials ~nd methods From 1973 Io lg86, we treated 430patients with squamous cell carcinoma of the mobiIe tongue by if'radiation, Staging was carried out according to the 1979 UICC 6assifieadon. In this group, wc had 147 T 2 N o: 70 were treated by braehyth~apy alone for
the primary site ~ d 77 others with combined external beam i~adiarion and braebylherapy. All the patients were followed for a mthinlum of 3 ye~s. Four patients, who died of u n k n o ~ causes, were classified as dead from cancer. interstitial implants were perfomled ~ith iridium192, Many early patients wen treated with hairpins, but since 1980, we have used o~ly the plastic tube loop technique (Ftg 1). We think the optimal spacing between the radioactive lines is about 1,2-1,5 em (maximum 2 em). Computerized dosimotry was generated according to the prinelpJes of the P~is system. The tumor volume was ~eompassed by the reference isodose (85~ of the basal dose) Since 1980, intraoral leaded shields have been consistently used to protect the mandible during treatment. In the braehytherapy Mone group, the dose applied on the reference isodose was within 88 G~ and 79 G) but in six cases, the dose was given on a m o ~ external
F;~ ) Tcc~nlque of lrldlum-192brachph~rz W wl,h #~tlo tub~s ~everalpos~ibils z ~ to the number and positionof th~ {~ps ~ the ~xte,u
of~h~ ,esion,
225
isodose, T h e dose rate ranged from 0.31 a n d 1.09 Gyfh, 31 patients h a d a neck d i s s ~ t i o n but only two r ~ e i v e d a n additional ~ t e ~ a l irradiation. 39 patients h a d n o treatment on the n o d e areas, P a t i ~ t s trealed with the combined t r e a t m ~ t r ~ e i v ed e x t e ~ a l b e a m i~ediafion to the primary tumoral site a n d to the nodal a ~ a s ~Sth O~ or 6 M e V X ra~s. W e uifiified p~allel opposed portals for the primary a n d superior n o d a l areas ~ l d a single anterior portal for the inferior ~ d supraclaaleular areas with protection ofifie vocal cords a n d o f tile spinal cord. External radiotherapy w a s given at 200 e G y per fraction ( I fififi c o y a week); all the fields were irradiated each day. F o r the primary tumor, the ~ t e ~ a l b e a m i~aifiataln dose w a s within 40 a n d 50 G~q but it w a s only 20, 35 a n d 36 G y in three cases. T h e braehyifierapy boost w a s 30.9 G y ( S D 5,1). T h e external b e a m i~adiation dose on the node areas w a s always within 40 a n d 5 0 G y Car these patients (but 36 G y in one case). T h e m e ~ age in the group treated with brachyifierapy alone o n T w a s 58.2 ( S D 11.5) compared to 59.5 (fiD Ifi.8) in the other group. Successful sab, age of ~ u r m n c e w a s defined as no evidence of disease for >~ 1 year following the salvage treatment. Compfieations consisting o f soft tissue necros~s or b o n e exposure were s e o ~ d according to their severity: g r a d e 1 . 2 or 3. This seems to correlate with the classification ofMendenhall et at. [8 ] as minor, moderate a n d severe complications. T h e computerized statistical study w a s m a d e with the Pigns program. T h e actuarial survival w a s calculated using Kaplan-Meier methodology a n d l o g - r ~ k test. R~ults
Local and I~o~gional control Local ~ d I~orcgional control did not change a t the 5 ~ d 10 year l a t e r a l s sifiee n o r ~ n c e was obse~ed a h e r 3 years. Local control w a s higher for the patients treated wiifi brachytherapy alone versus combined braehyifierapy a n d e x t e m a l b e a m r 89,8 % versus 50.6% ( p ~ 0.00002) a n d 67.6% versus 46.5% for the Incaregional conlrol ( p - 0.029) (Taifie I), W e observed more n o d a l r e o u ~ c e s in the patients treated b y brachytherapy alone. I n 39 out o f 70 cases, n o nodal disseedon w ~ p e r f o ~ e d . E v e n in these cases, the difference in control is signifie~t. T h e analysis might h a v e been biased b y a slight predominance o f w o m e n in the brachytherapy group (19 versus 6 in the o t h ~ group). Nev~theless, the differ~ c e w a s alike for the males o f both groups (87ifi~ versus 47,3~o for local control) ( p - O.00O2) a n d
TABLE T2N ~ local and Iocoregqonal c~trol at $ ~e~s (All r e c ~ c ~
o~urre0 within 3 rea~ a ~ r treatmem). B~acb~h~apy ~e~s~ e~,~r~ ~ i ~ h ~ p y brachytherapy
IERT~ §
Implant alone (70 pin)
898
676
ERT + implant (77 pts)
506 p ~ 0.00a02
465 . ~ 0029
TABLE II
Implant alone
8g~2% (57 ptd)
91.6% 02 pls)
ERT § implanl
584% (48 pts) p ~ 0.0]l
37.2!/o ( ~ pin) p~00l[
Despite the ~ h o . . ~mb~ of I~i~.~ > 3 ~m i~ bo,h grn~0~ and t ~ various .umber of patiems in eachone, the dlfrerencr b~wcen 1hose two gcoups i~ alw~y~ highly significanL
76.l% versus 46.2% for locoregional control ( p - 0 . f i 0 5 ) . M o r e o v ~ in the bracbytherapy ifione
g ~ u p , m ~ showed b e t t ~ results t h a n w o m ~ (77% versus 50% for ]ocoregional control). Additionally, w e wondered whether the differen~ w a s not d u e to lesion size as there were slightly more lesions measuring between 3 a n d 4 c m in the group treated b y association t h a n in the brachyifierapy alone group, but the differences r ~ a i n e d significant in both groups w g h p - fi.01L despite the short n u m b e r o f p a d c n t s in each group a n d despite the difi'~ent numbers o f patients in both groups (Table II). T h e average tumoral surface in the patients for w h o m it w a s evaifiable a e c u r a t d y w a s d, 1 cm 2 in the first group ( S D 1.8) a n d 5.2 cm 2 in the s ~ o n d group ( S D 2.7).
Possible causes of recurren~ analyzed Group with brachytherapyalone technique w e h a d a few more r ~ u ~ n e e s w i t h the hairpin t ~ h nique t h a n with the plastic t u b e one. Perhaps one explanation is a difference in the tmgnt volume, although this difference is not significant with few recurrences in each group (Table II1). 56 pafien~s treated with doses betwee~ 6000 ~ d 7500 cGy. there were seven (7/56) r ~ u r r ~ e e s . Thene
226 TABLE Ill Rccurreaceg a c ~ ] n g to the treatment m Odolit{~s
TABLE V R~uw~ccs ~ T ~ M g
to the ERT dose.
T~ N~ (brach)therap 3 a{onr on Th r~urrenccs Reemre.ces ~ T
Compllcat{o~s~mae 2 aria 3
Hai~ins (27 pts)
5(185";)
~ a d e 2 = I (3,7~ )
pl~tig lube~ (43 pls}
2 (46%)
grado 2 - 6 (13.9~;) grade3 • i (23~1
were 0/12 recurrences in patients w h o r ~ e i v a d doses higher t h ~ 75 G y (but with a co~esponding increase in complications; T a b l e IV). F o r dose rates b e t w e ~ 40 a n d 100 r we found n o diff~ence in the rate o f ~cu~ences. T h e amount o f tumor margin is also very important. T h e s ~ u r i t y margin may he ~ p r e s s a d as the ratio between the treated volume a n d the estimated tumoral volume. This average ratio is less in p a O ~ t s with reeur~ n e e t h a n in those without r ~ u r r e n c e ( p is n o t significant because of the d{meulties to estimate the tumoral volume). In the group of patients with braehythcrapy alone, w e h a d n o more reeu~enees in the group o f tumors whose size w a s u n d e r 3 em than in the group of tumors w h o s e size w a s over 3 era. In the patients treated with br aehytherapy alone, w e noted one isolated n o d e recurrence in 4/31 p afients with neck dissection ~ d 13/39 patients without neck diss ~ t i o n I n the first group. 3 out of 4 patients with n o d e recu~ence died o f nodal evolution ~ d one of metas+ tases. In the second group, a m o n g the 13 patients with nodal recurrences, 3 h a d a tumor recurrence, over the s ~ e period, a n d only 4 o u t of 13 w e ~ cured with ~t salvage neck dissection (Table IV), I n the 77 patients treated with combined e x t ~ n a l b e a m i~sthation a n d brachytherapy, we o b s e ~ a d 37
Dose5 <4500 m y ~oo0c%,
26% 52%
(6~2~l (2~al~) ~ ooss
The rate of recurrences ~eems laver+ely related to extem~ therapy
primary reeu~ences. Salvage surgery w a s successful in six o f these cases, b u t three patients died secondary of t h t ~ e u ~ e m t h s ~ s e . In ,his group, w e noted that the local control w a s higher for p a t i ~ t s receivthg doses < 45O0 c G y t h ~ for pabents r ~ a l v i n g 50OO e G y with external b e a m i~adintloa ( p = 0.055); w e h a d very few patients mealving less t h a n 4000 c G y (Table Y).
Sur~iml fiu~ivals ~ e s h o w n in Figs. 2 a n d 3 for both ~ o u p s . T h e overall survival is 51.1% at 5 y e ~ s in the brach htherapy alone group versus 3 3 . 3 ~ in the combined external b e ~ a n d b r a c h y t h ~ a p y group (32.6 % versus 20.8% at 9 ye~s). Tile actuarial specific su~,ival* in the same groups, w a s 62,2% versus 34.7% at 5 years a n d 62.2% versus 2 8 . 2 ~ at 9 y e ~ s , respectively. T a b l e V l shows the causes o f death, either cancer or others ( s ~ o n d cancer, i n t ~ c u ~ e n t disease or unknown).
TABLE IV Recurrences according 1o dose. dose mtu and neck alssection.
Dose rate: > 40 < 100 cGy/~ ~ no difl~r~ee Dose ~+0,;7S oy ~lzs~(71~)
>750~,
! *
,
~
.
,
s
+
7
a
9
= 0 Z (all 2) (complications}
Neck dissection 4' = 11% (41311~ " ~ =0oo+ " • ERT ifdeck ~{,mo{on ~ posM~e
Fi~ 2 sp~l~o ~..~l~.lforbo,h ~oups l~el~a{~ patlo,,,~d e~d or mm~c~m di~ea~ or s+eo~d c~.coO
+ spedoc s.~J~{: w { m o m m * p~,t+,t+who died or im+,~c~,
TABLE V]H
Compl[~/on~ accordingle dose and dosera/~ Bracifi~ h~apy nIonc: complic atlo~s (N~d* 2 or 3)
t)o~, i~os) ~ 6000~ 7500: >7C~0 : Dose rnte {c~,~ ) ~50 : >50 :
':I- ~ 7 ~ 7 ~
~"
.
.
.
.
Fig-30~erall sur~dvalfor both groups.
Distributionof the c~use or death (%) among all deaths o~both
alone
~rae~ythcr~p~
Locore#ona/cancer
21y~ (8)
4~% (27)
io
~o% (1~}
~7% 116>
Com#i9 The complications were staged into three grades: grade 1 = mild complication; ~ade 2 ~ moderate eomptication; grade8 = seven~ complication. The incidence and severity of soft tissue n~rosis ~ d ] ~ bone exposure ~ e given th Tables VII and VIII. It appears that one patient out of five had a grade 1 complication (otter 6 to 8 months alter tream~ent). This set,is t~lcorrespond to the time of onset of devaseularization, In these cases, soeh limited ulcerations usua0y hea~ within 2 months with eonsewative management.
ThBLE VII T2No complications:~'ade 2 ~ 3 (moderateor sewei Implantalone EP.T + {replant
9% (s,'56) 25*. O,12) 0% {0~71 114~. 18163)
gcadc2 = 10~. (7/7Qi ~rade3= i.a% /l/~OI (norequiredsurgery) ~ndc 2 = 7.7~* (6/77l grade 3 = 3.8% i3"/77I
Seven moderate and one seve~ complication w ~ e observed in the braehytherapy group, including fiw
patients with bone e• (grade 0) and one with bone necrosis (grade 0), None of the patients required surgery. In the combined group, 4 moderate and 8 severe bone complications were observed (2 required surgery). There were only 2 spit tissue (grade 2) complications in this g~up. In the group treated sv~th brachytherapy alone we obse~ed increased complications with doses > 7500 cGs; the only grade 3 complicafion was in this group (Table VII). The pati~t s treated with inlerstitial implant alone sho~ing moderate or sev~e complications had received an average bracbyther spy dose of 7600 cGy. In the same group, the patients without complications had received an average dose of 6000 cGy (not statistically significant). In the combined treatment groap the average external radiation dose ~ d the average brachytherapy dose did not code[ate with complications. Diseusifion
Our data indicate mat, for T~ No lesions, the rates of ]oczl control, Iocore~ona] control and su~ival ~e significantly higher for patients treated with a single int~sthlifiimplant alone compared to the results of braebytherapy plus external beam irradiation. Differenees in doses provided to the tumor, respectively, by external irradiation and by brachytherapg may explain these outcomes. In our view, tile differ~ce in tumoral control between the patients treated with braehytherapy alone and those treated with combined external irradiation and brachyth~apy, may be due to the d0Tr in the dose p~vided, In br aehytherapy, the dose expressed is that on the 85 % isodose and all the points in the treated volume are applied higher doses (15 ~ higher, at least, for the furthest poims from the soure~ and much higher for the nearest ones). Moreover, the dose in brachytherapy is given continuously over a few days.
228 Other authors h a v e noted the same r a l a t i ~ s h i p b e t w e ~ the p~hability o f local controI a n d raodalibes o f trea~rtent. M a z e m n [I,3,7] noted a higher rate o f control for patients treated with i r i d i ~ i m p l ~ t alone a n d increasing rates of local contral vath inereaalng dose ( ~<6000 cGy, L C = 78 ~ ; ~>7000 Gy, L C = 9 4 ~ for T~ T~ p a d ~ t g ) . Hale e t a l . [5], as well as Pemot et a l I 10], previously indicated (1083) that the results were belier for T, T~ treated exclusively wish braehytherapy, F u e~ aI. [4] noted a he{tar control for T z lesions treated ~ t h h r a c h y t h ~ a p y alone compared with those treated by combined extc~al b e a m radiotherapy a n d h r ~ h y t b e r a p y . In thh series reported b y Hofiuchi el at. [6], local control with b r a e h ~ h e r a p y alone (radium or gold grains) w a s superior for T I a n d T z presentations t h a n with the eombalation. H e holed that " p r i m a r y c o n t ~ l rates ~Jth the combined treatment o f e x t ~ n a l h e ~ a n d interstitial treatment for infiltrative T : a n d / o r T~ lesions ~ e ~ apparenlly Io~er t h a n those for the group ~eeiving imalants alone".
More r~ently, Mendenhal[ et at. [8,9] indicate that "for T 2 lesions, the rate of local control is slgnificandy higher for patients w h o ~ e i v e at least half the dose ~ t h a n int~sdtial implant" A s a boost t ~ h n i q u e , W ~ g [11] prefers the intraoral cone with electrons, but lhe interstitial implants in his series ~ r e radium implants a n d not comparahle with iddalm-192 implants. However. boosl h y intraoral eo~e m a y h e a va!~d a i l e r ~ l ] v e whel~ d is ~eehnically posalblc, Conclusion G i v ~ the previous rcsalts a n d those oftheliterature, w e think it would be better to treat cancers o f the mobile tongue T~ Nu b y bracbythcrapy only for the primary lesion, however, a systentatic treatment must h e adjuncted on the node areas, namely a n ~ k diss~tion or systematic prophylactic neck irradiation, in order to avoid recurrences.
Ref~en~s I Besk, V. M~ernn, J ]., Grimard, L , Crook, 1, Hadd~d, ~, Piedbnis. p. cal~tehL E., Raynel, m., Martin. M , Lehoargcois I P. and e~r ~ O)mparison of curietherapy vcr~u~ ext~nat [rtad~ati~ ~mbined ~ith eurietherapy in stage it
squ~mou~ edl ca~inomas of the m~bile tongue. Radtoth=r Oneul, I$: 339-347,1990. 2 Chu, A andFIctebcr, G.H. Incidencesandeausesoffailn~to ~ n t ~ [ by irradiation the primary Iesion in squamous cell carcinomas of the ~terior two thirds of the tonic and fl~r of mouth A m J Ro~tg~nol liT: 502-558.1973, 3 Cr~k, J ,Mazerou,JJ,,Marincllo,G.,Walap, W. andI~ierquin, B Prognostic factors of local outcome for TI ~2 carcinomas of oral tongue Ireat~d by Iridium 192 implantati~, "[he Creleit experience ]nt J Radial Oucol. Biol Phys 17 (Sappl I): ]?0-17L ]989. 4 Fv, K.K,Chan, E.K,Phillips, T.L.andRayJ, W Time.do~ and volume factor~in inzecsfitJalradiumimalams of earemom~
of the ~ral tongue Radiology ] 19: 209-213, IgTfi. 5 Hale, C, Gcrbaalet, A , Wibault, P,, Chaessgne, D ~ d Marand~. p, p,~sultat8 de In r el de I'assoeiation radioth6ra01r tr auseut an6e~uriet h~raple darts ] 55 eaac~s de la langne mobile. Exp~rien~ de l'Instimt Oustave gou~sy de t970 197g, in: Cancers de ]a Langue; A~ualit~s de Carelnologie Cervieof~ei~le, pp. 53 57, Masson, p~is, 1983. 6 HoriuchL J,,Okuyama, T., Shibnya, H. and Takeda, M. Results
ofb=aeh:her~py fo, c~m~, orthc ,ungu~ ~ ( h ~p~ial ~mrl,~i~ on local pro~osls Int I R~diaL O~col Biol. Phys. 8: fl29-835, 1982 7 Mazerou, J, J,, Crook, 1, B~k. V , Walop, W, afld Pierquin, B. Irid]mn 192 implamatlon for T1 T2 epidermoid earclnom~ of the mobile tongue: The Crelr experience, Int. J. Radiat. Oncol BioL phys. 17: (Snppl. [l; 225,1959. 8 Mendenhall, w . M , Parson, i, T., Stmng~, s. P., cz.~sisi, N. J, and Million, R, R. Tz oraho~ae eareinoma trested ,~ith radiotherapy: analysis of local c~lrol and ~mplieati~s, Rad~other. ennui. 16: 275-282,1989, 9 Meudenhal], W. M , Vml C~e. W S, Bova, F.i. and Million, g.X.A.alysis oni~e d~se r~c,~s in squamons een e.~einoma of the oral t~gue and floor of meuff~ ~mated wKh ra~Oon therapy a~one i m i Radial. On~ol. 3iol rhys 7; 1005-1011. lO eemohM.,Malisssrd. L.,Camla~d. M Bey,g , Holl~leltcnS, Scou~acher, P. a.d Kogm[nski,p Carr162 ~pldermoldes
de la langue mobile. Quelle ~nd~itr th&~peutlque adopter? A propos de 22I eas trait~s au Cent{e Alexis Vaurin entre 1972 et 1979, In: Caners de Ia L~ngue, Aetua[it~s de Csreinologie ~rvico-Fac~Nc, ~ 147-156 M ~ o a , paris, 1983 I1 Wang, c . c . Radlotherapcudr management and restdts of TI NO, T2 NO carcinoma of the ~ral tongue: ewluation of boost techniques. Int J. Radiat Oneol, BIoLph~s, 17:297-29 L 19~9.