Search

The first part of our work has focused on the relationship between tumor volume and tumor control. Indeed, it is well known that the importance of irradiated volume could be a main parameter of radiation-induced complications. Numerous mathematical models have described the correlation between the irradiated volume and the risk of adverse effects. These models should predict the complication rate of each treatment planning. At the present time late effects have been the most studied. In this report we firstly propose a review of different mathematical models described for volume effect. Secondly, we will discuss whether these theoretical considerations can influence our view of radiation treatment planning optimization.

Volume is an important parameter of radiation therapy. Local control is inversely related to tumor size and the complication rate increases with the importance of the irradiated volume. Although the effect of irradiated volume has been widely reported since the beginning of radiotherapy, it has been less studied than other radiation parameters such as dose, fractionation, or treatment duration. One of the first organ system in which the adverse effect of increased volume was well defined is the skin. Over the last twenty years, numerous mathematical models have been developed for different organs. In this report we will discuss the relation between irradiated volume and tumor control. In a second article we will study the impact of irradiated volume on radiation adverse effects.

Resume Objectif de l’etude. – Etude retrospective de 23 cas de cholangiocarcinome de la voie biliaire principale traites par une irradiation externe conformationnelle. Patients et methodes. – Huit patients ont ete irradies apres une resection radicale (R0), en raison de l’atteinte des ganglions (N+) ; sept apres une resection microscopiquement incomplete (R1) et huit en l’absence d’exerese (R2). Il a ete delivre une dose de 45 a 50 Gy chez tous les patients. L’irradiation a ete completee dans un volume reduit jusqu’a 60 Gy en cas d’exerese incomplete (R1) ou d’absence d’exerese (R2). Une chimiotherapie concomitante a ete administree dans 15 cas. Resultats. – Un patient a souffert d’une stenose du duodenum et un autre de l’anastomose biliodigestive. Deux patients sont decedes d’une angiocholite recidivante, sans qu’il soit possible d’affirmer une rechute tumorale. Cinq patients etaient en situation de remission complete, six ont eu une rechute locale, quatre une carcinose peritoneale et six une dissemination metastatique osseuse ou viscerale. Les taux de survie actuarielle etaient respectivement de 75, 28 et 7 % a un, trois et cinq ans (duree mediane de survie : 16,5 mois). Sept patients etaient en vie avec un recul de quatre a 70 mois. La survie est comparable dans les trois groupes. Les resultats decevants dans le groupe R0N+ (trois rechutes locales et trois carcinoses peritoneales) pourraient etre lies a l’absence de complement d’irradiation jusqu’a 60 Gy. Les rechutes etaient le plus souvent a distance apres exerese incomplete (R1) et locales et peritoneales en l’absence d’exerese (R2). Conclusion. – Une chimioradiotherapie conformationnelle de 60 Gy represente une approche palliative interessante pour les carcinomes de la voie biliaire principale localement evolues.

To evaluate the complications and oncologic and functional results of preoperative radiochemotherapy and sphincter-saving resection for T3 cancers of the lower third of the rectum.Carcinomas of the lower third of the rectum (i.e., located at or below 6 cm from the anal verge) are usually treated by abdominoperineal resection, especially for T3 lesions. Few data are available evaluating concomitant chemotherapy with preoperative radiotherapy for increasing sphincter-saving resection in low rectal cancer.Between 1995 and 1999, 43 patients underwent preoperative radiochemotherapy with conservative surgery for a low rectal tumor located a mean of 4.5 cm from the anal verge (range 2-6); 70% of the lesions were less than 2 cm from the anal sphincter. There were 40 T3 and 3 T4 tumors. Patients received preoperative radiotherapy with a mean dose of 50 Gy (range 40-54) and concomitant chemotherapy with 5-FU in continuous infusion (n = 36) or bolus (n = 7). Sphincter- saving resection was performed 6 weeks after treatment, in 25 patients by using intersphincteric resection. Coloanal anastomoses were associated with a colonic pouch in 86% of the patients, and all patients had a protecting stoma.There were no deaths related to preoperative radiochemotherapy and surgery. Acute toxicity was mainly due to diarrhea, with 54% of grade 1 to 2. Four anastomotic fistulas and two pelvic hematomas occurred; all patients but one had closure of the stoma. Distal and radial surgical margins were respectively 23 +/- 8 mm (range 10-40) and 8 +/- 4 mm (range 1-20) and were negative in 98% of the patients. Downstaging (pT0-2N0) was observed in 42% of the patients (18/43) and was associated with a greater radial margin (10 vs. 6 mm; P =.02). After a median follow-up of 30 months, the rate of local recurrence was 2% (1/43), and four patients had distal metastases. Overall and disease-free survival rates were both 85% at 3 years. Functional results were good (Kirwan continence I, II) in 79% of the available patients (n = 37). They were slightly altered by intersphincteric resection (57 vs. 75% of perfect continence; NS) but were significantly improved by a colonic pouch (74 vs. 16%; P =.01).These results suggest that preoperative radiochemotherapy allowed sphincter-saving resection to be performed with good local control and good functional results in patients with T3 low rectal cancers that would have required abdominoperineal resection in most instances.

Resume Les cancers de l’estomac peuvent se propager par voie hematogene, lymphatique et intraperitoneale. Ils interessent un organe de taille et de forme variables, entoure d’organes critiques parmi les plus vulnerables a la radiotherapie. Le volume tumoral macroscopique (GTV) est a considerer pour les irradiations preoperatoires ou exclusives. Apres une chirurgie non radicale, un GTV « residuel » doit etre defini avec l’aide du chirurgien. Le volume–cible anatomoclinique (CTV) doit inclure trois volumes intriques. a) Un volume d'extension tumorale. Les rechutes locoregionales apres une exerese apparemment complete sont aussi frequentes que les rechutes metastatiques. Le risque depend surtout de l'extension en profondeur dans la paroi gastrique et de l'atteinte ganglionnaire. L'infiltration de la paroi gastrique peut aller au-dela des limites macroscopiques de la tumeur, en particulier dans les « linites plastiques ». Ce volume devra donc inclure : la tumeur (ou la zone ou elle siegeait), ainsi que l'espace occupe par l'estomac, plus une partie du colon transverse, le duodenum, le pancreas et le tronc porte. En postoperatoire ce CTV inclura le montage anastomotique. b) Un volume d'extension peritoneale. Il est souhaitable de considerer deux stades d'extension peritoneale : (1) une extension microscopique par contiguite, en regard des tumeurs T3 et surtout T4, que l'on peut esperer steriliser avec des doses de 45 a 50 Gy ; (2) une « carcinose peritoneale », avec des nodules multiples ou situes a distance de la tumeur, qui ne releve plus d'une radiotherapie mais d'une chimiotherapie systemique ou intraperitoneale. Le CTV englobera donc la portion de cavite peritoneale contenant le segment gastrique atteint et si possible le peritoine parietal sous la cicatrice de laparotomie. c) Un volume d'extension lymphatique comprenant les groupes de 1 a 16 de la classification japonaise. Seule la partie superieure du territoire lomboaortique (groupe 16) sera incluse. Ce volume doit s'elargir pour englober le pedicule hepatique et le hile de la rate. En bas, le CTV pourra etre restreint au seul volume lymphatique, si la situation haute ou interne de la tumeur permet de proteger davantage d'intestin grele. Pour les tumeurs situees aux deux extremites de l’estomac, un envahissement microscopique des limites d’exerese – de pronostic defavorable – devra etre l’objet d’une surimpression. Pour les tumeurs du cardia , le CTV sera au moins etendu a la partie inferieure de l’œsophage thoracique et du mediastin posterieur. En cas d’atteinte du bas œsophage, chez un sujet jeune et peu denutri, il faudra inclure l’ensemble des territoires mediastinaux ou se drainent le bas œsophage et le cardia. Pour les tumeurs de la grosse tuberosite , la coupole sera incluse, ainsi que la rate et son hile (ou leur lit d’exerese). Si la tumeur ne s’etend pas a la petite courbure, on pourra etre plus econome au niveau du pedicule hepatique. Pour les tumeurs de l’antre , il faudra inclure le pedicule hepatique, les ganglions du ligament hepatoduodenal. En revanche, pour les tumeurs qui ne se prolongent pas le long de la grande courbure, la couverture du hile splenique semble inutile. En conclusion, le traitement de ces cancers demeure un exercice difficile et doit etre realise sous le controle de radiotherapeutes confirmes.

Resume But : Les adenocarcinomes de la jonction anorectale, en particulier les lesions T3, sont habituellement traites par une amputation abdominoperineale. Le but de cette etude etait d’evaluer les resultats oncologiques et fonctionnels apres traitement radiochirurgical conservateur de cancers de la jonction anorectale. Methodes : De 1990 a 1999, parmi 395 patients operes d’un cancer du rectum, 31 ont eu une chirurgie conservatrice pour une tumeur situee entre 2 et 4,5 cm (moyenne : 3,6) de la marge anale. Il y avait 16 hommes et 15 femmes, d’âge moyen 62 ans (extremes : 30–86). Les lesions etaient reparties de la facon suivante : 5 T2, 23 T3 et 3 T4 ; 17 etaient N1 et 3 M1. Une radiotherapie preoperatoire a ete realisee chez 26 patients (dose 46 Gy, extremes : 36–54), en association avec une chimiotherapie concomitante dans 14 cas. Six semaines plus tard, une resection rectale etait realisee, enlevant une partie ou la totalite du sphincter interne. Les anastomoses colo-anales etaient associees a un reservoir colique en J dans 22 cas et a une stomie de protection dans tous les cas. Resultats : Il n’y a pas eu de mortalite postoperatoire. Sept complications sont survenues : trois fistules anastomotiques, trois hemorragies pelviennes, une pancreatite aigue. Trois patients ont garde une stomie definitive. Apres radiotherapie, un down staging (pT0-2 N0) a ete observe dans 46 % des cas (12/26). La marge de resection distale etait de 2,2 cm (extremes : 1–3) microscopiquement saine dans tous les cas. La marge de resection laterale etait non envahie (⩾1 mm) dans 97 % des cas. Avec un recul de 36 mois, aucune recidive locale n’a ete suspectee. Vingt-six patients (84 %) etaient vivants, 23 sans recidive. La moitie des patients avaient une continence normale, l’autre moitie des fuites mineures occasionnelles. Les resultats fonctionnels etaient meilleurs avec un reservoir colique. Conclusion : Le traitement conservateur des adenocarcinomes de la jonction anorectale est possible sans compromettre le controle pelvien et la survie des patients. La radiotherapie preoperatoire et la resection intersphincterienne ont permis d’obtenir des marges de resection distales et laterales saines.

Background After abdominoperineal resection (APR), anorectal reconstruction with dynamic graciloplasty has been proposed to avoid abdominal colostomy and improve quality of life. Graciloplasties involving one or two gracilis muscles with various configurations have been described. The aim of this study was to evaluate morbidity and functional results in a homogeneous series of patients undergoing double dynamic graciloplasty following APR for rectal cancer. Patients and methods From May 1995 to May 1998, 15 patients (ten men and five women; mean age 54 (range 39–77) years) underwent anorectal reconstruction with double dynamic graciloplasty after APR for low rectal carcinoma. All patients had preoperative radiotherapy (45 Gy), 11 with concomitant chemotherapy, eight had intraoperative radiotherapy (15 Gy) and ten received adjuvant chemotherapy for 6 months. The surgical procedure was performed in three stages: APR with coloperineal anastomosis and double graciloplasty (double muscle wrap); implantation of the stimulator 2 months later; and ileostomy closure after a training period. Results There was no operative death. At a mean of 28 (range 3–48) months of follow-up, there was no local recurrence; two patients had lung metastases. Early and late morbidity occurred in 11 patients, mainly related to the neosphincter (12 of 16 complications). The main complication was stenosis of the neosphincter (n = 6), which developed with electrical stimulation. Of 12 patients available for functional outcome, seven were continent, two were incontinent and three had an abdominal colostomy (two for incontinence, one for sepsis). Compared with patients without stenosis, patients with neosphincter stenosis required major reoperations (four versus zero) and had a poor outcome (two of six versus five of six with a good result). Conclusion The double dynamic graciloplasty is associated with a high risk of neosphincter stenosis, which may entail morbidity, reintervention and poor functional results. The stenosis is a heterogeneous feature of the neosphincter induced by asymmetrical traction of both gracilis muscles. It is suggested that single dynamic graciloplasty should be used for anorectal reconstruction after APR.

Resume Introduction L'adenocarcinome du pancreas exocrine represente un probleme de sante publique, avec 6000 morts chaque annee en France. Actualites et points forts Le facteur de risque le moins discute est l'intoxication tabagique. Ce cancer est le plus souvent diagnostique a un stade avance. Les resultats de la chirurgie radicale restent mediocres. Dans les meilleures series, moins d'un patient sur quatre survit plus de cinq ans. Perspectives et projets La radiochimiotherapie postoperatoire augmente un peu l'espoir de guerison. Dans les tumeurs localement evoluees, la radiochimiotherapie, exclusive ou preoperatoire permet d'obtenir quelques survies au-dela de deux ans. Si les resultats de la chimiotherapie palliative restent mediocres, cette arme ne doit pas etre negligee car elle peut apporter un benefice clinique en preservant mieux et plus longtemps la qualite de vie de certains patients. Cela a ete montre dans deux essais controles comparant le seul traitement symptomatique et son association a la chimiotherapie. Perspectives et projets La disponibilite de nouvelles molecules peut permettre d'esperer des modalites plus performantes d'association radiochimiotherapiques dans les cancers localement avances et d'aboutir a des protocoles de polychimiotherapies plus efficaces dans les cancers metastases.

Resume Objectif de l'etude Evaluer retrospectivement les resultats a long terme de l'irradiation fractionnee des neurinomes de l'angle pontocerebelleux. Methodes et materiel De janvier 1986 a octobre 1995, 29 patients ayant un neurinome de l'angle pontocerebelleux de stade III ou IV ont ete traites par irradiation externe fractionnee. Un patient a ete consecutivement traite des deux cotes. Les indications ont ete les suivantes: contreindications generales a la chirurgie (16 patients); preservation de l'audition apres exerese chirurgicale controlaterale (six patients); exerese chirurgicale incomplete (cinq patients); recidive non chirurgicale (trois patients). Une technique par trois ou quatre champs fixes focalises et coplanaires a ete utilisee. Les doses ont ete prescrites sur les isodoses 95 ou 98 %; tous les champs ont ete traites chaque jour, cinq jours sur sept, pour une dose totale mediane de 51 Gy (1,8 Gy par fraction). Les photons de 6 a 10 MV ont ete utilises dans la majorite des cas. Resultats La mediane de la duree de la surveillance etait de 66 mois (extremes: 7–120 mois). Sept patients sont decedes, deux de progression tumorale, cinq de cause non tumorale. Une reduction du volume tumoral et/ou l'apparition d'une necrose centrotumorale ont ete observees chez 13 patients (43,3 %); pour 14 patients (46,6 %), l'image radiologique est restee stable. Enfin, le volume tumoral a progresse chez trois patients. Deux patients ont beneficie d'une exerese chirurgicale complete apres la radiotherapie (une tumeur stable et une tumeur en progression). L'audition a pu etre preservee chez quatre des six patients atteints de neurinomes bilateraux. Aucune neuropathie faciale ni trigeminee n'a ete observee. Conclusion Cette experience confirme que l'irradiation fractionnee des neurinomes de l'angle pontocerebelleux de grande taille est efficace et bien toleree. L'audition peut etre maintenue.

PURPOSE: Retrospective analysis of 17 patients with intracranial germ cell tumors treated in a multidisciplinary consultation at the Bordeaux University Hospital a and literature review. MATERIALS AND METHODS: Seventeen consecutive patients were treated from 1978 to 1995 for a primary intracranial germ cell tumor. Median age was 14 (range 3-29 years). There were two malignant teratoma, six proved germinoma and nine presumed germinoma (diagnostic based on biological, radiological and treatment criteria). All received radiotherapy from 30 to 60 Gy (median 40 Gy) in different volumes. Chemotherapy was administered in 15 cases, three after surgery and 12 after radiotherapy. RESULTS: All tumours were in complete remission after initial treatment. The two malignant teratomas recurred in non-irradiated area after nine and 48 months, and the patients died. None of the germinoma recurred within a follow-up period of two to 17 years (median 65 months). Five and 10 year actuarial overall survival rates were the same: 84% for all histologies and 100% for germinomas. Only two patients developed school difficulties and six presented an hypopituitarism, of which one was consecutive to radiotherapy. Chemotherapy was well tolerated. CONCLUSION: This retrospective study and literature analysis are in favor of limited dose and volume of radiation therapy associated with chemotherapy.

Resume Objectif de l'etude Evaluer les resultats a long terme de l'irradiation fractionnee des meningiomes intracrâniens. Patients et methodes De janvier 1981 a decembre 1996, 156 patients atteints d'un meningiome intracrânien ont ete traites par irradiation externe fractionnee. L'âge median etait de 57 ans. Les indications ont ete les suivantes: histologie agressive apres exerese chirurgicale complete (12 patients); exerese chirurgicale incomplete, quelle que soit l'histologie (37 patients); risque fonctionnel du fait du site tumoral (base du crâne) et/ou contre-indications generales a la chirurgie (77 patients); recidive, reoperee ou non (30 patients). Une technique par trois ou quatre faisceaux fixes focalises et coplanaires a ete utilisee. Depuis 1993, une dosimetrie en trois dimensions apres simulation virtuelle a ete realisee pour 71 patients. Les doses ont ete prescrites sur les isodoses 95 ou 98 %; tous les champs ont ete traites chaque jour, cinq jours sur sept, pour une dose totale mediane de 50 Gy (1,8 Gy par fraction). Des photons de 6 a 10 MV ont ete utilises dans la majorite des cas. Resultats La duree mediane de la surveillance a ete de 40 mois. La tolerance immediate a ete excellente avec une amelioration clinique en cours de traitement pour 19 patients (17,8 %). L'evaluation clinique a long terme a montre une stabilisation ou une remission des symptomes pour 130 patients (83,4 %). Radiologiquement, 124 tumeurs ont ete controlees (79,4 %), dix ont progresse dans le volume irradie, cinq en bordure; six patients ont eu une nouvelle localisation a distance. Les taux de survie actuarielle globale et specifique etaient respectivement de 75 et 89 % a cinq ans, et de 45 et 76 % a dix ans. Conclusion Cette experience confirme que l'irradiation fractionnee des meningiomes intracrâniens est efficace et generalement bien toleree a moyen et a long termes. L'analyse des recidives a confirme l'importance d'une bonne definition du volume cible et engage a discuter de la dose totale a delivrer.

The purpose of this research was to reevaluate long-term results of frationated radiation therapy (RT) in two previously published series of cerebello-pontine angle (CPA) neurinomas. From January 1986 to October 1995, 29 patients with stage III and IV CPA neurinomas were treated with external fractionated RT. One patient was irradiated on both sides and indications for radiotherapy were as follows: (a) poor general condition or old age contraindicating surgery, 16 cases; (b) hearing preservation in bilateral neurinomas after contro-lateral tumor removal, 6 cases; (c) partial resection or high risk of recurrence after subsequent surgery for relapse, 5 cases; (d) nonsurgical relapse, 3 cases. Most patients were irradiated with 6 to 10 MV photons. A three- to four-field technique with coplanar static beams and conformal blocks was used. Doses were calculated on a 95% isodose and were given 5 days a week for a mean total dose of 51 Gy (1.80 Gy/fraction). Median follow-up from RT was 66 months (7 to 120); 7 patients died, 2 with progressive disease, 5 from nontumoral causes. Two patients underwent total tumor removal after RT (1 stable and 1 growing tumor). On the whole, tumor shrinkage was observed in 13 patients (43.3%), stable disease in 14 (46.6%), and tumor progression in 3. Hearing was maintained in 4 out of 6 hearing patients. No patient experienced facial or trigeminal neuropathy. Long-term efficacy of fractionated RT is well documented in this series. Acute and delayed tolerance was excellent. Hearing can be preserved for a long time.

Resume But de l'etude Evaluer de facon prospective une technique de simulation virtuelle (SV). Patients et methodes De septembre 1993 a fevrier 1997, nous avons mis en traitement par cette technique 343 patients. Les sieges tumoraux les plus frequents etaient le systeme nerveux central (132), le rectum (59), les bronches (43) et la prostate (28). Une tomodensitometrie a ete realisee en position de traitement, avec 25 a 70 coupes jointives debordant largement le volume a irradier. Les contours du volume cible (CTV de l'ICRU 50) et des organes critiques ont ete traces coupe par coupe par le radiotherapeute. Les faisceaux couvrant le CTV plus une marge de securite (PTV) ont ete mis en place. Les cliches virtuels de simulation (DRR) ont ete reconstruits. Ils ont permis de verifier que le PTV etait bien couvert et d'affiner le plan d'irradiation. L'isocentre definitif a ete mis en place au simulateur classique, les cliches reels devant correspondre aux cliches virtuels. Resultats Nous avons ainsi observe des effets secondaires marques (grade 3) dans 19 cas et severes (grade 4) dans un cas. Avec un recul median de 18 mois, cinq complications de grade 3 sont apparues et une de grade 4. Cinquante-cinq cancers ont rechute dans le volume irradie et 19 en bordure. Conclusion Appliquee aux tumeurs profondes, la SV est devenue chez nous une technique de routine. Sa realisation demande environ 2 heures au radiotherapeute. Elle stimule la reflexion sur les volumes a traiter, l'anatomie et les voies d'extension des tumeurs. Elle represente un bon outil pedagogique.

To reevaluate long-term results of fractionated radiation therapy (RT) in a previously published series of cerebello-pontine angle neurinomas (CPA).From January 1986 to May 1992, 24 patients with Stage III and IV CPA neurinomas were treated with external fractionated RT; 7 patients had phacomatosis. One patient was irradiated on both sides and indications for radiotherapy were as follows: (a) poor general condition or old age contraindicating surgery, 14 cases; (b) hearing preservation in bilateral neurinomas after contralateral tumor removal, 5 cases; (c) partial resection or high risk of recurrence after subsequent surgery for relapse, 4 cases; (d) nonsurgical relapse, 2 cases. Most patients were irradiated with 9 MV photons. A three- to four-field technique with coned-down portals was used. Doses were calculated on a 95% isodose and were given 5 days a week for a mean total dose of 51 Gy (1.80 Gy/fraction).Median follow-up from RT was 60 months (7 to 84); five patients died, two with progressive disease. Two patients underwent total tumor removal after RT (one stable and one growing tumor). On the whole, tumor shrinkage was observed in 9 patients (36%), stable disease in 13 (52%), and tumor progression in 3. Hearing was maintained in 3 out of 5 hearing patients with phacomatosis.Fractionated RT appears to be an effective and well-tolerated treatment for Stage III and IV CPA neurinomas. Hearing can be preserved for a long time.

Resume Parmi les criteres de qualite essentiels d'une irradiation externe, la reproductibilite du traitement est un element majeur: cours de chaque seance d'irradiation, la meme dose doit etre delivree tres precisement dans le meme volume. Les systemes d'acquisition automatique d'images digitalisees des champs d'irradiation permettent a l'heure actuelle, d'apprecier et de corriger ≪au jour le jour≫ les erreurs de positionnement des champs d'irradiation, avant meme la poursuite du traitement. Dans cette etude nous utilisons le systeme portai vision varian (PVV) au cours d'une irradiation pelvienne de routine.Dans les premieres secondes de chaque seance d'irradiation, des images de chaque champ traite en photons de 18 MV sont prises chez 16 patients. Sur chaque image, les anomalies observees sont aussitot classees par le radiotherapeute selon leur type (deplacement tete-pieds, lateral ou anteroposterieur) et leur gravite (deplacement necessitant ou pas une remise en place). En cas de deplacement en dehors de la marge de securite technique, le patient est remis en place avant poursuite de la seance. Au total, 60% des champs traites ont ete analyses au PVV soit 922 champs pelviens. Parmi eux 901 (97,7%) sont interpretables. Deux cent quatre-vingt-douze anomalies de positionnement sont detectees (33%) dont 59 (6%) imposent a nos yeux une remise en place immediate. Les anomalies sont le plus souvent observees sur les champs anteroposterieurs (35% contre 29,5%) et correspondent surtout a des deplacements lateraux ou anteroposterieurs. Finalement, chaque patient presente, au cours de son traitement, plus de 30% d'anomalies (en moyenne 18 a 19 anomalies par patient). Ce sont huit fois sur dix des deplacements minimes. Cependant, 14 des 16 malades etudies (88%) presentent des erreurs necessitant correction. Leur nombre moyen est limite a quatre par patient. L'etude du morphotype des patients ne nous a pas permis d'isoler une population a risque. Ceux traites avec les plus grands champs semblent pourtant avoir un taux d'erreur plus important. En conclusion, nous pensons qu'a l'heure actuelle au cours d'une irradiation pelvienne de routine, l'utilisation quotidienne d'un systeme de type portal vision est la condition sine qua non d'un traitement de qualite. En effet, cette nouvelle technique permet de decouvrir en temps reel chez la plupart des patients (14/16), un nombre consequent d'erreurs (en moyenne quatre par patient) necessitant une remise en place du patient avant poursuite de l'irradiation.

From January 1986 to March 1989, 20 patients with stage III and IV cerebello-pontine angle neurinomas were treated with external fractionated radiation therapy; seven patients had phacomatosis. Indications for radiation therapy were as follows: (a) poor general condition or old age contraindicating surgery, 10 patients; (b) hearing preservation in bilateral neurinomas after contro-lateral tumor removal, 5 patients; (c) partial resection or high risk of recurrence after subsequent surgery for relapse, 4 patients; (d) non-surgical relapse, 1 patient. Most patients were irradiated with a 9 MV linear accelerator. A 3 to 4-field technique with 5 × 5 cm portals was used. Doses were calculated on a 95% isodose and were given 5 days a week for a mean total dose of 5140 cGy (180 cGy/fraction). Median follow-up from radiation therapy was 30 months (7 to 46); 4 patients died, 2 with progressive disease. Two patients underwent total tumor removal after radiation therapy (1 stable and 1 growing tumor). On the whole, 14 tumors remained stable, 3 decreased in size, and 3 progressed. CT scan or NMR tumor changes consistent with partial tumor necrosis appeared in four cases. Hearing preservation was obtained in 35 hearing patients with phacomatosis. When surgery is not indicated or incomplete, fractionated radiation therapy appears to be an effective and well-tolerated treatment for stage III and IV neurinomas. Hearing can be preserved.

The first part of our work has focused on the relationship between tumor volume and tumor control. Indeed, it is well known that the importance of irradiated volume could be a main parameter of radiation-induced complications. Numerous mathematical models have described the correlation between the irradiated volume and the risk of adverse effects. These models should predict the complication rate of each treatment planning. At the present time late effects have been the most studied. In this report we firstly propose a review of different mathematical models described for volume effect. Secondly, we will discuss whether these theoretical considerations can influence our view of radiation treatment planning optimization.

Volume is an important parameter of radiation therapy. Local control is inversely related to tumor size and the complication rate increases with the importance of the irradiated volume. Although the effect of irradiated volume has been widely reported since the beginning of radiotherapy, it has been less studied than other radiation parameters such as dose, fractionation, or treatment duration. One of the first organ system in which the adverse effect of increased volume was well defined is the skin. Over the last twenty years, numerous mathematical models have been developed for different organs. In this report we will discuss the relation between irradiated volume and tumor control. In a second article we will study the impact of irradiated volume on radiation adverse effects.

Retrospective study of 23 patients treated with conformal radiotherapy for a locally advanced bile duct carcinoma.Eight cases were irradiated after a radical resection (R0), because they were N+; seven after microscopically incomplete resection (R1); seven were not resected (R2). A dose of 45 of 50 Gy was delivered, followed by a boost up to 60 Gy in R1 and R2 groups. Concomitant chemotherapy was given in 15 cases.Late toxicity included a stenosis of the duodenum, and one of the biliary anastomosis. Two patients died from cholangitis, the mechanism of which remains unclear. Five patients are in complete remission, six had a local relapse, four developed a peritoneal carcinosis, and six distant metastases. Actuarial survival rate is 75%, 28% and 7% at 1, 3 and 5 years, respectively (median: 16.5 months). Seven patients are still alive with a 4 to 70 months follow-up. Survival is similar in the 3 small subgroups. The poor local control among R0N+ cases might be related to the absence of a boost to the "tumor bed". In R1 patients, relapses were mainly distant metastases, whereas local and peritoneal recurrences predominated in R2.Conformal radiochemotherapy delivering 60 Gy represents a valuable palliative approach in locally advanced biliary carcinoma.

The spread of gastric adenocarcinoma may follow three main patterns: hematogenic, lymphatic and intraperitoneal. A GTV should be considered in preoperative or exclusive radiation therapy. After non-radical surgery, a "residual GTV" will be defined with the help of the surgeon. The CTV encompasses three intricated volumes. a) A "tumor bed" volume. After radical surgery, local recurrences appear as frequent as distant metastases. The risk depends upon the depth of parietal invasion and the nodal status. Parietal infiltration may extend beyond macroscopic limits of the tumor, especially in "linitis plastica". Therefore this volume will include: the tumor and the remaining stomach or their "bed of resection", a part of the transverse colon, the duodenum, the pancreas and the truncus of the portal vein. In postoperative RT, this CTV also includes the jejuno-gastric or jejuno-esophageal anastomosis. b) A peritoneal volume. For practical purposes, two degrees of spread must be considered: (1) contiguous microscopic extension from deeply invasive T3 and T4 tumors, that remain amenable to local sterilization with doses of 45-50 Gy, delivered in a CTV including the peritoneal cavity at the level of the gastric bed, and under the parietal incision; (2) true "peritoneal carcinomatosis", with widespread seeds, where chemotherapy (systemic or intraperitoneal) is more appropriate. c) A lymphatic volume including the lymph node groups 1 to 16 of the Japanese classification. This volume must encompass the hepatic pedicle and the splenic hilum. In proximal tumors, it is possible to restrict the lower part of the CTV to the lymphatic volume, and therefore to avoid irradiation of large intestinal and renal volumes. In distal and proximal tumors, involvement of resection margins is of poor prognosis--a radiation boost must be delivered at this level. The CTV in tumors of the cardia should encompass the lower part of the thoracic esophagus and the corresponding posterior mediastinum. In tumors invading the distal esophagus, a more complete coverage of mediastinal lymph nodes should be considered, especially in patients in good general condition. In tumors of the gastric fundus, most of the left hemidiaphragm should be included, as well as the spleen and its hilum (or their resection bed). In proximal tumors without involvement of the lesser curvature, a full coverage of the hepatic pedicle is not necessary. In contrast, for distal tumors, the hepatic pedicle and the hepatoduodenal ligament should be included whereas the splenic area could be spared.Planning the treatment of gastric cancer remains difficult; target volumes must be customized by experienced radiation oncologists according to tumoral and clinical situation.

Deep venous thrombosis is 50 times less frequent in upper than in lower limbs. Data remain poor in the literature. Forty consecutive patients (24 males, 16 females, mean age: 54.5 years) were retrospectively analysed from 161 subjects who underwent venous explorations of the upper extremity for a 3.5 year period in the same center. Diagnosis of thrombosis was made by duplex ultrasonography (n =37) or phlebography (n =3). Main clinical manifestations were edema (n =36) and pain (n =29). Location of thrombosis was humeral (n =1), axillary (n =2), or sub-clavian (n =37, 2 bilateral). The majority of thrombosis (n =29) were secondary to cancer and venous catheter (n =19, 15 implanted ports), to central catheter alone (n =3) or cancer alone (n =7). The 11 others were associated with thoracic outlet syndrome (n =6) or apparent primary thrombosis (n =5). Thrombophilia was identified in 6 out of these 11. During follow up [mean of 9 months (0,5-36)], two patients developed pulmonary embolism, 14 a post-thrombotic syndrome and 16 patients died. Initial therapy included heparin (n =36) or fibrinolysis (n =4). Upper extremity deep venous thrombosis are mostly associated with cancers and venous catheters. Thrombophilia is frequent in the other cases. Heparin followed by oral anticoagulation is the optimal therapy whose duration depends upon underlying condition. Fibrinolysis has not been useful for preventing post-thrombotic syndrome in our study.

Adenocarcinomas of the anorectal junction, especially T3 lesions, are usually treated by abdominoperineal resection. The aim of this study was to evaluate oncologic and functional results following conservative radiosurgical treatment of cancers of the anorectal junction.From 1990 to 1999, among 395 patients with rectal carcinoma, 31 had sphincter-saving resection for a tumour located between 2 to 4.5 cm (mean 3.6) from the anal verge. There were 16 men and 15 women, mean age 62 years (range 30-86). There were 5 T2, 23 T3 and 3 T4 tumours; 17 were N1 and 3 were M1. Preoperative radiotherapy was performed in 26 patients (dose: 46 Gy, range: 36-54), with concomitant chemotherapy in 14 cases. Intersphincteric resection was performed six weeks after neoadjuvant treatment. Coloanal anastomoses were associated with a colonic pouch in 22 cases and with a protecting stoma in all cases.There was no postoperative mortality. Seven complications occurred: 3 anastomotic fistulas, 3 pelvic haemorrhages and 1 acute pancreatitis. Three patients had a definitive stoma. After preoperative radiotherapy, down-staging (pT0-2 N0) occurred in 46% of cases (12/26). Distal margin was 2.2 cm (range: 1-3) and was microscopically safe in all cases. Lateral margin was safe (or = 1 mm) in 97% of cases. With a mean follow-up of 36 months, no local recurrence was suspected. Twenty-six patients (84%) were alive, 23 free of disease. Half of the patients had perfect continence, whereas the other half had occasional minor soiling. Functional results were better in patients with a colonic pouch.Conservative treatment of carcinomas of the anorectal junction is possible without compromising pelvic control and patient survival. Pelvic control was probably achieved by using preoperative radiotherapy with intersphincteric resection, ensuring safe distal and lateral margins.

Pancreatic carcinoma is a major public health concern, as it kills more than 6,000 people each year in France.The main risk factor demonstrated by concordant case-control studies is cigarette smoking. Pancreatic carcinoma is generally diagnosed at an advanced stage. Results of radical surgery are still poor. In most of the reported series, less than 25% of the patients survive at five years.Postoperative radiochemotherapy slightly increases the hope of cure. In locally advanced tumors, radiochemotherapy, sometimes preoperative, allows some patients to survive more than two years. Though results of palliative chemotherapy remain very poor, some clinical benefit has been observed in randomized trials comparing this treatment with the currently best supportive treatment.

PURPOSE: Standard surgical treatment for low rectal cancer situated below 5 cm from the anal verge or at less than 1 cm from the anal ring is abdominoperineal resection. This is because of the necessity both to achieve a sufficient distal margin and to preserve the whole of the anal sphincter. The aim of this study was to evaluate morbidity, oncologic, and functional results of intersphincteric resection with excision of the internal anal sphincter and low coloanal anastomosis for carcinomas of the anorectal junction. METHODS: From January 1990 to December 1996, 16 patients were studied prospectively. All patients had an infiltrating adenocarcinoma (5 T2 and 11 T3), located between 2.5 and 4.5 (mean, 3.6) cm from the anal verge. Rectal resection with a minimum distal margin of 2 (mean, 2.4) cm was performed in all cases; six patients underwent partial resection of the internal sphincter, and ten patients had a subtotal resection. A colonic J-pouch was associated with coloanal anastomoses in eight cases. Twelve patients had preoperative radiotherapy, 3 with concomitant chemotherapy; 5 patients had postoperative chemotherapy. RESULTS: There was no post-operative mortality. Morbidity occurred in four patients, of whom two underwent permanent colostomy after pelvic hemorrhage or anovaginal fistula. After a median follow-up of 44 (range, 11–92) months, no local recurrence was observed, and two patients died of distal metastases. The five-year actuarial survival rate was 75 percent. Continence was normal in one-half of patients and was altered in the other patients who suffered from occasional minor leaks. The median resting pressure was lower after subtotal than after partial resection of the internal sphincter (40vs. 70 cm H2O;P=0.02), but functional results were similar in the two groups. CONCLUSION: These preliminary results suggest that intersphincteric resection can be an alternative to abdominoperineal resection for selected rectal tumors situated at the anorectal junction, without compromising chance of cure. Functional results and continence were not altered by subtotal resection of the internal anal sphincter.

To evaluate the long-term results of external fractionated radiation therapy (RT) in the treatment of intracranial meningiomas.From January 1981 to December 1996, 156 patients with intracranial meningiomas were treated with external fractionated RT. Median age was 57. Indications for radiation therapy were as follows: (1) completely excised histologically aggressive tumors (12 patients); (2) incomplete surgical resection (37 patients); (3) medically inoperable or basilar tumors where operation would involve considerable danger or permanent neurological damage (77 patients); and, (4) tumor recurrences (30 patients). Most patients were irradiated with 6 to 9 MV photon beams. A three to four-field technique with coned-down portals was used. Since 1993, 71 patients had a three dimensional dosimetry. Doses were calculated on the 95% or 98% isodoses, all fields were treated every day, five days a week, for a median total dose of 50 Gy (1.8 Gy/Fraction).Median follow-up from radiation therapy was 40 months. Acute tolerance was excellent; an early clinical improvement during radiation therapy was noted in 19 patients (17.8%). Clinical improvement or stabilization was observed in 130 patients (83.4%). Radiologically, local control was obtained in 124 cases (79.4%) and tumor recurrences occurred in 21 cases (ten progressions in the treated volume, five borderline, six new locations). Overall and cause specific-survival rates were 75% and 89% at five years, and 45 and 76% at 10 years, respectively.These results reassess the role of fractionated RT in the treatment of intracranial meningiomas. Long-term tolerance is excellent for a majority of patients. The study of recurrences confirms the importance of the definition of the target volume, and asks questions about total given doses.

To evaluate retrospectively the long-term results of fractionated radiation therapy (RT) in cerebello-pontine angle neurinomas (CPA).From January 1986 to October 1995, 29 patients with stage III and IV neurinomas were treated with external fractionated RT. One patient was irradiated on both sides and indications for RT were as follows: (1) general contraindications for surgery (16 patients); (2) hearing preservation in bilateral neurinomas after controlateral tumor exeresis (six patients); (3) partial tumor removal (five patients); and, (4) non-surgical recurrence (three patients). A three to four fields technique with coplanar static beams and conformal cerobend blocks was used; doses were calculated on a 95 to 98% isodoses and were given five days a week for a median total dose of 51 Gy (1.8 Gy/fraction). Most patients were irradiated with 6 to 10 MV photons).Median follow-up was 66 months (seven to 120 months). Seven patients died, two with progressive disease, five from non-tumoral causes. Tumor shrinkage was observed in 13 patients (43.3%), stable disease in 14 (46.6%), and tumor progression in three. Two patients underwent total tumor removal after RT (one stable and one growing tumor). Hearing was preserved in four out of six patients. No patient experienced facial or trigeminal neuropathy.Fractionated RT is a well tolerated and efficacious treatment of large non-surgical CPA neurinomas.

Retrospective analysis of 17 patients with intracranial germ cell tumors treated in a multidisciplinary consultation at the Bordeaux University Hospital a and literature review.Seventeen consecutive patients were treated from 1978 to 1995 for a primary intracranial germ cell tumor. Median age was 14 (range 3-29 years). There were two malignant teratoma, six proved germinoma and nine presumed germinoma (diagnostic based on biological, radiological and treatment criteria). All received radiotherapy from 30 to 60 Gy (median 40 Gy) in different volumes. Chemotherapy was administered in 15 cases, three after surgery and 12 after radiotherapy.All tumours were in complete remission after initial treatment. The two malignant teratomas recurred in non-irradiated area after nine and 48 months, and the patients died. None of the germinoma recurred within a follow-up period of two to 17 years (median 65 months). Five and 10 year actuarial overall survival rates were the same: 84% for all histologies and 100% for germinomas. Only two patients developed school difficulties and six presented an hypopituitarism, of which one was consecutive to radiotherapy. Chemotherapy was well tolerated.This retrospective study and literature analysis are in favor of limited dose and volume of radiation therapy associated with chemotherapy.

The aim of this study was to assess the technical and functional results of total anorectal reconstruction with double dynamic graciloplasty after abdominoperineal resection (APR).From May 1995 to December 1996. 10 patients (6 males and 4 females), with a mean age of 54 years (range 39-74), underwent anorectal reconstruction for low rectal adenocarcinoma. All patients had preoperative radiotherapy and six had postoperative chemotherapy. The surgical procedure was performed in three stages: 1) APR, coloperineal anastomosis, double graciloplasty and ileostomy; 2) three months later, implantation of stimulator and leads; 3) after a two-months training period, the stoma was closed.There was no postoperative mortality. Early and late morbidity occurred in 5 patients: 2 colonic fistulas, 1 necrosis of colon, 1 ileostomy prolapse, 1 neosphincter stenosis, 1 sepsis of stimulator. No patient had recurrence of the disease (mean follow-up 16 months), but two patients died at 3 and 8 months, respectively from anorexia and pulmonary embolism. Seven patients were available for evaluation (2 fistula, 1 death). Before training, the resting pressure and the squeeze pressure were 30 and 175 cm H2O respectively. At the time of evaluation, the electrical stimulated pressure was 95 cm H2O. Six of these 7 patients were continent (5 with spontaneous defecation, 1 with enemas) and 1 was incontinent.Anorectal reconstruction with dynamic graciloplasty can be an alternative to permanent colostomy for selected patients after APR. However, there is a high morbidity and the quality of life of the patients must be evaluated.

Prospective evaluation of a virtual simulation technique.From September 1993 to February 1997, 343 patients underwent radiation therapy using this technique. Treated sites were mostly: brain (132), rectum (59), lung (43), and prostate (28). A CT-scan was performed on a patient in treatment position. Twenty-five to 70 jointive slices widely encompassed the treated volume. The target volume (CTV according to ICRU 50) and often critical organs were controured, slice by slice, by the radiation oncologist. Beams covering the CTV plus a security margin (PTV) were placed on the "virtual patient". Digital radiographs were reconstructed (DRR) as simulator radiographs for each field. Thus, the good coverage of PTV was assessed. Fields and beam arrangements were further optimized. Definitive isocenter was then placed using a classical simulator. Perfect matching of DRR and actual simulator radiographs had to be obtained.Nineteen patients presented grade 3, and 1 grade 4 acute radiation effects. With a median follow-up of 18 months, five patients suffered from grade 3, and one from grade 4 complications. Fifty-five patients had tumor recurrence in the treated volume, and 19 had marginal relapse.In our department, virtual simulation has become a routine technique of treatment planning for deep-seated tumors. This technique remains time-consuming for radiation oncologists: about 2 hours. But it stimulates reflexion on anatomy, tumor extension pathways, target volumes; and is becoming an excellent pedagogical tool.

Purpose : To evaluate efficacy and tolerance of external fractionated radiation therapy (RT) in the treatment of intracranial meningiomas. Methods and Materials : From January 1981 to September 1993, 91 patients with intracranial meningiomas were treated with fractionated RT. Indications were as follows: (a) incomplete surgical resection, 20 patients; (b) tumor recurrences without considering the amount of the second resection, if performed, 14 patients; (c) completly excised angioblastic, aggressive benign, and anaplastic tumors, 8 patients; (d) medically inoperable and basilar tumors where operation would involve considerable danger or permanent neurological damage, 44 patients. Most patients were irradiated with 6 to 9 MV photon beams. A three- to four-field technique with coned-down portals was used. Doses were calculated on the 95% isodose and were given 5 days a week for a median total dose of 52 Gy (1.80 Gy/fraction). Results : Median follow-up from radiation therapy was 40 months. Acute tolerance was excellent, but there were six late delayed injuries. Tumor recurrences occurred in six cases. Six patients died from their tumor or RT complications, 19 from nontumoral reasons. Three, 5- and 10-year survival rates were 82, 71, and 40%, respectively. The most significant prognostic factor was age: 5-year survival rate was 86% for patients less than 65 years and 37% for patients more than 65. However, there were no differences in recurrence-free survival rates between patients younger than 65 and the oldest ones. Of 60 symptomatic patients with neurological deficits, 43 had neurological improvement (72%), beginning in some cases within 15 to 20 days after starting RT. Conclusion : These results reassess the role of fractionated RT in the treatment of meningiomas, and stress on its efficacy, especially on cranial nerves palsies, without severe toxicity in most cases.

From May 1988 to June 1992, 129 eligible patients suffering from measurable advanced colorectal cancer were enrolled in a randomized study comparing bolus fluorouracil plus leucovorin (FU-FA); continuous fluorouracil infusion (FU-cont); FUcont plus cyclophosphamide and mitomycin C (FUMIC). FU-FA consisted of weekly fluorouracil (FUra) bolus (600 mg/m2) 1 hour after the initiation of a 2-hour infusion of 500 mg/m2 of leucovorin, for 6 weeks every 8 weeks. FUcont patients were planned to receive 400 mg/m2/day FUra infusion, for 21 days every 28 days. In FUMIC patients, FUcont was associated with weekly cyclophosphamide bolus (300 mg/m2) and monthly mitomycin C bolus (10 mg/m2). Quality of life was evaluated using six linear analogue scales, completed by the patient. Accrual in the FUMIC arm was stopped after the 25th patient because of toxicity. The response rates were 22 of 48 (45.8%) with FUcont and 13 of 52 (25%) with FU-FA (P = .048). Progression-free survival (median: 8 v 4.4 months; P = .0026) and overall survival (median: 12.9 v 9.6 months; P = .028) were significantly greater for the FUcont arm compared with the FU-FA arm. Toxicity was observed in 62% of the FUcont patients (grade 3-4: 10%), mainly hand-foot syndrome, diarrhea, mucositis, and mainly gastrointestinal in 69% of the FU-FA patients (grade 3-4: 11.6%). Linear analogue scales exploring quality of life, available for the first 6 months, gave similar scores in FU-FA and FUcont patients. We conclude that this FUcont schedule, achieving high FUra dose-intensity, offers significant advantages, in terms of response and survival, over weekly FUra plus leucovorin.

Treatment reproducibility is a major criterion of quality assurance in radiation therapy. During each course, the same dose should be delivered in the same volume of irradiation. Today, portal imaging devices can be used routinely to check and correct patient positioning before much of the daily irradiation has been delivered. In this study we used the Portal Vision Varian (PVV) system during pelvic irradiation in 16 patients. This device can automatically acquire portal images in the first seconds of each course. Observed discrepancies are directly classified by the radiation oncologist according to their type (cranio-caudal, lateral, antero-posterior) and severity (correction of patient positioning is necessary or not). In case of error, patient positioning is corrected before the end of irradiation. Of the 922 portals analysed with PVV, 901 could be analysed (97%). Two hundred and ninety-nine positioning discrepancies were observed (33%) with 59 of them leading to correction (6%). Most of the time, these errors concerned antero-posterior portals. Finally, each patient had an average of 18 to 19 discrepancies which were mainly of no importance for treatment quality. Nevertheless, real errors leading to correction were observed in 14 patients (88%) with an average of four per patient. In some patients many errors occurred, while in others only a few. These shifts were not related to patient weight and thickness but probably a portal dimension. In summary, we think that during pelvic irradiation a portal imaging device should be used daily to improve treatment quality. This system can help the radiation oncologist to discover many positioning errors (an average of four) in the majority of patients (88%) and to correct them before the end of irradiation.

For 5 years, 121 general practitioners (GP) had a one-week course of continuing medical education, organized by the League against cancer, by groups of five, in oncology university services. A few years later they were asked, through a postal questionnaire, about their appreciation of this course and its consequences on their practice. Seventy six (55%) answers indicate that, except for two GP, such a course was considered as very valuable. They think it was useful to improve their clinical expertise and to reduce radiological and biological investigations. Their relations with patients, families and oncologists had been improved. They hope such an experiment will continue. Limits and difficulties of this experiment are briefly underlined.

Since recent treatment planning systems calculate volumetric dose distribution, an objective evaluation of potential toxicity in the main critical organs may be helpful in treatment optimization. Modeling the toxicity of radiotherapy must at least account for: (a) specific risks in every critical organ; (b) total dose and dose per fraction; (c) partial irradiation of critical organs; (d) heterogeneous dose distribution. The Radiation Damage Factor formula is aimed at estimating the delayed toxicity of a given treatment plan on every critical organ concerned. The formulation uses a double exponential function: RDF=100 e −Ke −(a+bd)DVC c , where: D is the total dose, and d the dose per fraction; a and b are coefficients representing the radiosensitivity of the critical organ, according to the linear-quadratic model, with ab=a/β. K represents the theoretical critical unit content of the organ, these critical units being groups of functionally related stem cells. The avoidance of a complication depends on the ability of surviving critical units to preserve organ function. V is the ratio: irradiated volume/total volume of the organ. Exponent c accounts for tissue organization: c is equal to or near 1 in "parallel organs" like the liver or the lung, where localized hot spots are tolerated; c is lower in "series organs" like the spinal cord where hot spots, even in a small portion, are dangerous. Heterogeneous irradiation, summarized by dose cumulative-volume histograms, is accounted for by calculating step by step the dose D′ considered as having an equivalent effect when given in the largest irradiated volume ratio. Preliminary calibration of the RDF formula is attempted for radiation myelitis and radiation hepatitis.

The combined use of surgery and radiotherapy is commonly accepted as the most effective treatment for locally advanced head and neck cancers. T3 and T4 tumors of the oral cavity and oropharynx often necessitate extensive local surgery. From 1981 to 1988, 199 patients with T3 and T4 tumors of the oral cavity and oropharynx were treated. One hundred seventeen patients underwent surgery plus postoperative radiotherapy; 78 had flap reconstructions. This series is extremely homogeneous because surgery was always performed by two surgeons, whereas radiotherapy was the responsibility of the same physician. The results of this study show a 96% local control rate at the end of treatment among the patients with combined treatment. The average time by which hospitalization was prolonged due to surgery was 29 days. The type and delay of recurrences and survival in relation with node involvement are also discussed. Extensive surgery in association with radiotherapy remains a reliable treatment in such patients.

Fifty consecutive children with cerebellar medulloblastoma were operated on May, 1971 to December, 1988. At 5 and 10-years, actuarial survival rates were respectively 46%42%. To determine prognostic factors, various clinical and histological features were analysed: multivariate analysis showed that increased vascularity, coagulative necrosis, and postoperative clinical status have a significant influence survival. When necrosis and/or increased vascularity were present (H+), the 5-year survival rate was 17%, whereas it was 70% when absent (H-) (p less than 0.0001). Postoperative clinical status was described as poor when neurological symptoms such as vigilance or vegetative disorders, akinesia, hypotonia, apragmatism, mutism and emaciation were associated; the 5-year survival rate was 10% in the presence of such deficits (C+) and 55% in their absence (C-) (p = 0.0002). From these results, 2 distinct groups were identified: Group 1: 23 patients without pejorative factors (H- and C-); 5 and 10-year survival rates were 80%. Group 2: 27 patients with one or two pejorative factors (H+ and/or C+); 5 and 10-year survival rates were respectively 18% and 12% (p. less than 0.0001). From our experience, it appears that histology and postoperative clinical status can be early predictors of patient outcome, since neurosurgical and radiation therapy techniques have been largely optimized. These factors could be the basis for developing adapted treatment protocols.

Thirty adult patients treated for brain tumors by radiation therapy were tested by the Weschler intelligence scale (WAIS). All patients were in complete clinical and scanographic remission. Intellectual quotient (IQ) and deterioration coefficient were evaluated. This coefficient was determined by separate study of intellectual faculties according to their relationship to external aggressions. During the first six months, IQ and deterioration coefficient are similar to those observed in the normal population. From 7 to 30 months, there is a significant diminution of the intellectual faculties (p = 0,01) noted on the deterioration coefficient but without global IQ alteration. After 30 months appears a decrease in the global IQ, testifying of further alteration of intellectual faculties. Radiation therapy seems to be the predominant factor in the induction of this deterioration and the importance of the irradiation volume appears to be essential. On the other hand, neurosurgery, chemotherapy and emotional status seem to be less important. Intellectual deterioration makes professional reinsertion difficult but not impossible. Further studies are necessary to define what place take other factors like age, vascular status, tumor site, type and number of chemotherapy cycles, acute radiation tolerance, in the induction of this deterioration.

Computed tomography now is the primary method for the management of supratentorial malignant gliomas. In this study, analysis of CT scan controls done at 2 or 3 month intervals in 87 treated patients showed that: 1) in most instances, recurrence arose within the initial tumor site; 2) the tumor invades contiguous anatomical structures through intra-and inter-hemispheric association fibers; 3) multicentric gliomas are uncommon (less than 6% of patients). The purpose of this study was to reduce irradiated volumes and thus radiation toxicity. The finding that most patients experience recurrence within the original tumor site is evidence experience recurrence within the original tumor site is evidence that current radiation doses and total brain irradiation are inappropriate. Our results are consistent with those recently reported in the medical literature. Using computed tomography data, it should be possible to restrict the irradiated volume to the tumor site and to presumptive extension pathways. Higher doses may be delivered to this volume.

A statistical study was used to determine the significant prognostic factors in a protocol containing surgery, radiotherapy and polychemotherapy applied to 121 consecutive unselected patients suffering from malignant gliomas. The histological grade of the malignant tumor, the age of the patient and the neurological state if it could be evaluated after surgery appeared to be significant. It is therefore possible to suggest a clinical classification into three groups. Group 1 consists of patients aged 50 or less and a postoperative neurological state of 1 or 2 according to Order's classification; group 2 consists of patients with one pejorative factor, i.e. aged over 50 or a postoperative neurological state of 3 or 4; group 3 is those patients with two pejorative factors (aged over 50 and postoperative neurological state of 3 or 4) or aged 65 or over. Methods of application of treatment are also involved in the therapeutic response: Excision surgery appears to be essential. It improves the duration and quality of survival and influences the efficacy of combined radio-chemotherapy. By contrast, the effects of cobalt and chemotherapy must be envisaged in an efficacy/tolerance ratio which is difficult to evaluate, but for which the suggested clinical classification as a certain indicative value. Finally, the various factors lead to an approach to the concept of therapeutic responses and to determine within the total population of malignant gliomas various categories of patients in whom the profile is defined both by clinical characteristics and the tolerance of treatment and in whom different treatment techniques must be envisaged.

Neuropsychologic studies were performed in 49 adult patients with primary intracranial tumors following surgery and radiation therapy. In the absence of computed tomography (CT) scan evidence of tumor recurrence or neurologic disorders, patients' intellectual functioning was tested once (49 patients), twice (18 patients), or three times (5 patients) by the Wechsler Intelligence Scale (WAIS) between 1.5 and 110 months after the beginning of irradiation. Full scale IQ (FSIQ) and deterioration coefficient (DC) were evaluated. Time elapsed from treatment to initial evaluation (vertical study) determined three statistically different patient groups. During the first 4 months, DC and FSIQ did not differ from a normal population. From 5 to 30 months, there appeared a significant probability of intellectual deterioration (p = 0.01), whereas median FSIQ remained at a normal level. Beyond 30 months, deterioration disappeared in terms of DC, but median FSIQ fell (p = 0.01). Second and third evaluations (horizontal study) confirmed the initial evaluation. Nevertheless, later recovery from FSIQ impairment was seen in 4 of 5 patients who were less than 30 years old and in 2 patients between 30 and 50 years of age who returned to work early after treatment. We believe that early neuropsychologic testing to detect early psychologic impairment is needed in all adult patients treated for brain tumors, especially in younger patients. Psychologic assistance and an early return to work may be beneficial.

Compared with other cancers, supra-tentorial astrocytoma malignancy is not a result of metastatic diffusion but of local recurrence: this suggests that local treatment might be effective. Nevertheless, long term survivors do not exceed 5 to 10% at 3 to 5 years.