A nationwide audit of the use of radiotherapy for rectal cancer in Italy

A nationwide audit of the use of radiotherapy for rectal cancer in Italy
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  ORIGINAL ARTICLE A nationwide audit of the use of radiotherapy for rectal cancerin Italy G. Gagliardi  • S. Pucciarelli  • C. R. Asteria  • A. Infantino  • G. Romano  • B. Cola  • P. De Nardi  • M. Brulatti  • M. Lambertini  • E. Contessini-Avesani  • G. Casula  • C. Coco  • D. D’Amico  • F. F. Selvaggi  • C. Eccher  • G. D’Ambrosio  • F. Galeotti  • E. Jovine  • I. Demma  • A. Fianchini  • G. Ambrosino  • L. M. Casentino  • M. Fiorino Received: 12 November 2009/Accepted: 16 June 2010/Published online: 15 July 2010   Springer-Verlag 2010 Abstract  Background   There is good evidence that radiotherapy isbeneficial in advanced rectal cancer, but its application inItaly has not been investigated.  Methods  We conducted a nationwide survey amongmembers of the Italian Society of Colo-Rectal Surgery(SICCR) on the use of radiation therapy for rectal cancer inthe year 2005. Demographic, clinical and pathologic datawere retrospectively collected with an online database.Italy was geographically divided into 3 regions: north,center and south which included the islands. Hospitalsperforming 30 or more surgeries per year were consideredhigh volume. Factors related to radiotherapy delivery wereidentified with multivariate analysis.  Results  Of 108 centers, 44 (41%) responded to the audit.We collected data on 682 rectal cancer patients G. GagliardiColoproctology Unit, Ars Medica Hospital, Rome, ItalyS. PucciarelliDepartment Oncological and Surgical Sciences,2nd Surgical Clinic, Padua University, Veneto, ItalyC. R. AsteriaDepartment of Surgery and Orthopaedics, Azienda OspedalieraC. Poma, Mantua, ItalyA. InfantinoDepartment of Surgery, S. Maria dei Battuti Hospital,S.Vito al Tagliamento (Pn), ItalyG. RomanoDepartment of Colorectal Medical Oncology, National CancerInstitute of Naples, Fondazione G. Pascale, Naples, ItalyB. ColaDepartment of Surgery, AOU S.Orsola-Malpighi,Univeristy of Bologna, Bologna, ItalyP. De NardiDepartment of Surgery, San Raffaele Scientific Institute,University of Milan, Milan, ItalyM. BrulattiDepartment of Surgery, Bellaria Hospital, Bologna, ItalyM. LambertiniDepartment of Surgery, San Salvatore Hospital, Pesaro, ItalyE. Contessini-AvesaniDepartment of Surgery, IRCCS Fondazione Hospital Milan,Milan, ItalyG. CasulaDepartment of Surgery, Surgical Clinic, University of Cagliari,Cagliari, ItalyC. CocoDepartment of Surgery, Sacro Cuore Catholic Universityof Rome, Rome, ItalyD. D’AmicoGastroenterological and Surgical Department,1st Surgical Clinic, University of Padua, Veneto, ItalyF. F. SelvaggiDepartment of Surgery, Policlinico 2nd University of Naples,Naples, ItalyC. EccherDepartment of Surgery, S.Chiara Hospital, Trento, ItalyG. D’AmbrosioDepartment of Surgery, Civic Hospital,Lavagna Genova, ItalyF. GaleottiDepartment of Surgery, Civic Hospital, Rovigo, ItalyE. JovineDepartment of Surgery, Maggiore Hospital, Bologna, Italy  1 3 Tech Coloproctol (2010) 14:229–235DOI 10.1007/s10151-010-0597-9  corresponding to 58% of rectal cancers operated by SICCRmembers in 2005. Radiotherapy was used in 307/682(45.0%) patients. Preoperative radiotherapy was used in236/682 (34.6%), postoperative radiotherapy in 71/682(10.4%) cases and no radiotherapy in 375 (55.0%) cases.Ofthe236patientswhounderwentpreoperativeradiotherapy,only 24 (10.2%) received short-course radiotherapy, while212 (89.8%) received long-course radiotherapy. Of the 339stage II–III patients, 159 (47%) did not receive any radio-therapy. Radiotherapy was more frequently used in youngerpatients ( P \ 0.0001), in patients undergoing abdominoperi-nealresection(APR)( P \ 0.01)andinthenorthandcenterof Italy ( P \ 0.001). Preoperative radiotherapy was more fre-quentlyusedinyoungerpatients( P \ 0.001),inlargevolumecenters ( P \ 0.05), in patients undergoing APR ( P \ 0.005)and in the north–center of Italy ( P \ 0.05). Conclusion  Our study first identified a treatment disparityamong different geographic Italian regions. A more sys-tematic audit is needed to confirm these results and planadequate interventions. Keywords  Rectal cancer    Adjuvant radiotherapy   Neoadjuvant radiotherapy    Disparities Introduction Rectal cancer represents approximately 25% of gastroin-testinal malignancy with 49.000 cases diagnosed everyyear in the United States [1] and 15,000 in Italy [2]. In the past, local recurrence after surgery alone for rectal cancerwas reported to be as high as 30% [3].Standardization of surgical technique in the form of totalmesorectal excision and the use of radiotherapy withconcomitant 5-Fluorouracil- (5-FU)-based chemotherapyboth contributed to lowering the local recurrence rate inrectal cancer [4]. The use of neoadjuvant therapy in theform of short-course radiotherapy (2500 cGy/5 days fol-lowed by surgery) has been shown to reduce the localrecurrence rate compared to surgery alone or to selectivepostoperative adjuvant therapy [5]. The alternative to short-course preoperative radiotherapy is long-course preopera-tive chemoradiotherapy (CRT) consisting of 4500–5040 cGy and concomitant 5FU-based chemotherapy, fol-lowed by a waiting period of 6 to 8 weeks before surgery.The advantages of downstaging the tumor are associatedwith a lower local recurrence rate than postoperative CRT[6, 7]. Incorporation of radiotherapy for rectal cancer into clinical practice has been investigated in the United States[8], United Kingdom [9], France [10], Norway [11], Swe- den [12] and Denmark [13] but not in Italy. Therefore, we conducted a survey among surgicalcenters affiliated with the Italian Society of Colo-RectalSurgery (SICCR) on the use of adjuvant or neoadjuvantradiotherapy for rectal cancer.The primary aim of our study was to evaluate the use of radiation therapy by Italian colorectal surgeons, the timing(preoperative  vs.  postoperative) of radiation therapy, and,secondarily, to identify the factors affecting radiationdelivery. Materials and methods The survey was an retrospective, observational, multicenterstudy, including all patients with a new diagnosis of mid orlow rectal cancer (defined as rectal cancer located up to12 cm from the anal verge at rigid proctoscopy) whounderwent surgery between January 1st and December31st, 2005. All surgical centers affiliated with the ItalianSociety of Colorectal Surgery were contacted by email andinvited to participate on a voluntary basis. An onlinedatabase was created and devised to maintain patients’anonymity. Because of the retrospective nature of thesurvey, approval by an ethics committee was not required.To investigate whether geographic location played a role inradiotherapy delivery, Italy was subdivided into a northern,a central and a southern region, which included the islands.Hospitals were classified as either University or Commu-nity. Preoperative radiotherapy was administered aseither [ 4000 cGy with conventional fractionation, usually1.8 cGy/day, 5 days/week with or without concomitant5-FU-based chemotherapy followed by a waiting period of 6–10 weeks (long-course protocol) or as 2500 cGy/5 daysfollowed by immediate surgery. Postoperative radiotherapyor CRT was given with the same fractionation and dose of the long-course preoperative radiotherapy. The TNM Stage I. DemmaDepartment of Surgery, IRCCS Fondazione Hospital,Castellana Grotte Bari, ItalyA. FianchiniDepartment of Surgery, Surgical Clinic, University of Ancona,Ancona, ItalyG. AmbrosinoDepartment of Surgery, San Bortolo Hospital, Vicenza, ItalyL. M. CasentinoDepartment of Surgery, San Filippo Neri Hospital, Rome, ItalyM. FiorinoSICCR Science Center, Italian Society of Colo-Rectal Surgery,Rome, ItalyG. Gagliardi ( & )Department of Surgery, Tulane University School of Medicine,1430 Tulane Ave., SL-22, New Orleans, LA 70112-2699, USAe-mail: gagliarg@yahoo.com230 Tech Coloproctol (2010) 14:229–235  1 3  classification was used based on pathologic TNM accord-ing to the AICC TNM staging system [14]. Stage after localexcision or colostomy was classified as unknown. Becauseof the possible downstaging effect of preoperative irradi-ation, we did not try to correlate the pathologic stage withradiation delivery. Univariate analysis was conducted usingStudent’s t test for numerical variables and Fisher’s exacttest for categorical variables unless otherwise indicated.Stepwise logistic regression was used to screen for inde-pendent variables. Results Of 108 centers contacted, 44 (40.7%) participated.Information about 682 cases was collected. According tothe Italian Network of Cancer Registries, in 2005 theincidence of newly diagnosed rectal cancer in Italy wasapproximately 15,000 cases [2], and according to the2005 Society’s annual report, SICCR members operatedon 1,185 rectal cancers. Our sample, which is limited tocancers of the lower 2/3 of the rectum, included 57.5%(682/1185) of rectal cancers operated by SICCR membersin 2005.Radiotherapy was used in 307/682 (45.0%) patients.Preoperative radiotherapy was used in 236/682 (34.6%)patients, postoperative radiotherapy in 71/682 (10.4%) andno radiotherapy in 375 (55.0%). Of the 236 patients whounderwent preoperative radiotherapy, only 24 (10.2%)received short-course radiotherapy. Of 212 patients whounderwent preoperative long-course radiotherapy, all but10 (4.7%) were given CRT. Of 71 patients who underwentpostoperative radiotherapy, 7 (9.8%) did not receive con-comitant chemotherapy. Of 375 patients who did notreceive radiotherapy, 99 underwent postoperative chemo-therapy only.Of 682 patients, 50 (7%) had metastatic disease, and in43 (6%) the stage was unknown (local excision,  n  =  37;colostomy only,  n  =  6). All 37 patients with stage 0 dis-ease were treated with long-course preoperative CRT,making the rate of pathological complete response 17.5%(37/212). Of the 339 stage II–III patients, 159 (47%) didnot receive any radiotherapy. None of the stage IV patientsreceived radiotherapy.The surgical procedures performed are shown inTable 1. Surgeons were able to carry out sphincter savingsurgery in 579/682 (84.9%) patients. Even after excludingall Hartmann’s procedures, a permanent stoma was avoidedin 561/682 (82.7%) cases.The mortality rate was 2% (14/682), and major periop-erative complications occurred in 137/682 (20%) patients(Table 2). Anastomotic leaks occurred in 79/560 (14%)patients after primary anastomosis.The results of univariate analysis for factors associatedwith radiation delivery are shown in Table 3. Mean agewas significantly lower in patients receiving radiotherapy(63  ±  11 SD) than in patients not receiving radiotherapy(68.61  ±  11.59) ( P \ 0.0001). Patients treated in southernItaly and the islands were significantly less likely to receiveradiotherapy than patients treated in northern (OR 0.3549;95% CI 0.2198–0.5729;  P \ 0.0001) and in central Italy(OR 0.4819; 95% CI 0.2955–0.7857;  P  =  0.0033). Afterexcluding the patients who underwent local excision orcolostomy only, those who underwent APR were morelikely to receive radiotherapy than patients who underwentsphincter saving procedures (OR 1.815; 95% CI 1.184–2.782;  P  =  0.0068).Multivariate analysis (Table 4) determined that geo-graphic location, age and sphincter saving surgery wereindependent predictors of radiotherapy delivery. Table 2  Major perioperative complicationsNumber (%)Anastomotic leak 79 14 b Wound infection 36 5.3Pelvic abscess 23 a 3.2Death 14 2.0Bowel obstruction 10 1.5Bleeding 7 1.0Anastomotic stricture 4 0.6DVT/pulmonary embolism 4 0.6Pneumonia 3 0.4Rectovaginal fistula 2 0.3Myocardial infarction 2 0.3Ureteric injury 2 0.3Respiratory failure 2 0.3Gastric perforation 1 0.1 a Including one patients after local excision b Percentage referred to the 560 patients with primary anastomosis Table 1  Surgical proceduresNumber (%)Low anterior resection 501 73.5Abdominoperineal resection 100 14.7Local excision 37 5.4Coloanal anastomosis 18 2.6Hartmann’s resection 15 2.2Colostomy only 6 0.9Proctocolectomy with ileostomy 2 0.3Proctocolectomy with ileal-pouch 1 0.1Colectomy with ileo-rectal anastomosis 1 0.1Pelvic exenteration 1 0.1Tech Coloproctol (2010) 14:229–235 231  1 3  The results of univariate analysis for factors associatedwith preoperative radiation delivery are shown in Table 5.Mean age was significantly lower in patients who under-went preoperative radiotherapy (62.22  ±  11.56 SD) than inothers (68  ±  11) ( P \ 0.0001). Patients treated in southernItaly and the islands were less likely to undergo preoper-ative radiotherapy than patients treated in northern Italy(OR 0.5218; 95% CI 0.3114 to 0.8743;  P  =  0.0155) or incentral Italy (OR 0.4225; 95% CI 0.2498 to 0.7148; P  =  0.0010). Patients treated in centers with a low casevolume were less likely to receive preoperative radiation(OR 0.7153; 95% CI 0.5206 to 0.9830;  P  =  0.0440).Finally, patients who underwent an APR were more likelyto receive preoperative radiotherapy than patients whounderwent a sphincter saving procedure (OR 1.815; 95%CI 1.184–2.782;  P  =  0.0068).Multivariate analysis (Table 6) determined that geo-graphic location, age, case volume and sphincter savingsurgery were all independent predictors of preoperativeradiotherapy delivery. Discussion Our survey shows that a high number (47%) of patientswith advanced rectal cancer (stage II–III) did not receiveradiotherapy. This is higher than in a survey conducted inthe year 2000 in US where 35% of patients with advancedstage rectal cancer did not receive radiotherapy [8].Our univariate and multivariate analysis offers impor-tant clues for the reason why a substantial number of patients who might benefit from radiation therapy did notreceive it. The influence of older age on radiation deliveryis well recognized [15–17] and was confirmed by our study. Patients over 75 years of age are more than 30% of rectal cancer patients [16], and their exclusion from trialson the sole basis of age has been debated since other papersshow that the majority of elderly are capable of toleratingadjuvant therapy and benefit from it [18]. There was ahigher chance of receiving radiation therapy for patientsundergoing an APR compared to patients undergoingsphincter sparing surgery. Patients undergoing APR werealso more likely to receive preoperative CRT. We specu-late that this difference is due to an attempt at downstaging Table 3  Univariate analysis:factors associated with radiationdelivery  RT   radiotherapy a Risk related to 1 year aging b North vs. Center, Center vs.South and North vs. South c Includes all other sphinctersaving abdominal procedures d Includes all sphincterremoving surgeryVariable Received RTnumber (% or  ±  SD)Did not receive RTnumber (% or  ±  SD)OR (95%CI)  P  valueAge 62.7 ( ± 11.2) 68.6 ( ± 11.6) 0.956 (0.942–0.969)  \ 0.0001 a SexMale 179 (47) 203 (53) 1.185 (0.874–1.607) 0.2749Female 128 (43) 172 (57)GeographicNorth 163 (51) 157 (49) 1.284 (0.923–1.786) 0.1376 b LocationCenter 114 (45) 141 (55) 2.075 (1.273–3.383) 0.0034South 30 (28) 77 (72) 2.665 (1.656–4.287)  \ 0.0001Center case \ 30 139 (42) 190 (68) 0.806 (0.595–1.090) 0.1613Volume  C 30 168 (48) 185 (52)Community Hospital 199 (47) 228 (53) 1.188 (0.869–1.624) 0.2805University Hospital 108 (42) 147 (58)APR d 60 (58) 43 (42) 1.815 (1.183–2.786 0.0063LAR c 233 (43) 303 (57)Laparoscopic 49 (47) 55 (53) 1.064 (0.699–1.620) 0.7720Open 247 (46) 295 (54)No complication 243 (46) 302 (54) 0.917 (0.630–1.337) 0.6538Complications 64 (47) 73 (53) Table 4  Multivariate logistic regression analysis with stepwise pro-cedure: factors associated with radiation deliveryEffect Odd ratio estimatesCharacteristic Estimate (95% CI)  P  valueGeographiclocationNorth vs. South 2.883 (1.742–4.770)  \ 0.0001Center vs.South2.551 (1.514–4.300) 0.0004Age 1 year increase 0.952 (0.938–0.967)  \ 0.0001Sphincter saving APR vs. LAR 1.866 (1.184–2.941) 0.0071232 Tech Coloproctol (2010) 14:229–235  1 3  lower rectal tumors in order to perform a sphincter sparingsurgery as well as to the higher risk of positive circum-ferential margin [19].An interesting result of our survey is the difference of radiotherapy delivery observed between different Italiangeographic areas. Patients from southern Italy and theislands were significantly less likely to receive radiother-apy. This possibly reflects the disparity in access toradiotherapy facility observed between different Italianregions [20] or regional cultural differences. A geographicvariation in radiotherapy delivery has also been reported inthe US, France and England and has been attributed tosocioeconomic factors and to regional cultural variations[10, 16, 21]. Racial disparity in receiving adjuvant treat- ment and difference in the outcome for colorectal cancerhas been the focus of epidemiologic studies in the UnitedStates [22–24]; however, recent data show that socioeco- nomic status is a more important predictor of adjuvanttherapy use in colorectal cancer than race/ethnicity [25].Our study has limitations. A 40% response rate amongcolorectal surgeons is low, and limiting the survey tocolorectal surgeons may overestimate the use of radio-therapy. Furthermore, there was an uneven number of participating centers between the different Italian regions.Nevertheless, the difference observed between Italianregions is concerning in the light of recent data reporting anincrease in incidence and mortality from colorectal cancerin the Southern Italy [26]. Another interesting factoremerging from our survey is that the likelihood of under-going neoadjuvant therapy was higher in high-volumecenters. In defining high-volume centers, we have adopteda previously used cut-off of 30 patients [27]. In the liter-ature, there is no consensus on the number of cases percenter per year to define high-volume centers, the numbersvary from 6 to 50 [28]. In a study limited to elderlypatients, Rogers et al. already observed that high-volumecenters were more likely to use radiotherapy for rectalcancer but did not separate pre from postoperative [29].Our findings are similar to those of a German audit andprobably reflect that indication and delivery of neoadjuvanttherapy is logistically easier in high-volume centers wherepatients are assessed by a multidisciplinary team [30] andradiotherapy facility are more accessible. Some have pro-posed to identify a subgroup of high-risk patients (maleswith low lying cancers) that are more likely to benefit from Table 5  Univariate analysis:factors associated withpreoperative radiotherapyadministration  pRT   preoperative radiotherapy a Risk related to 1 year aging b North vs. Center, Center vs.South and North vs. South c Includes all other sphinctersaving abdominal procedures d Includes all sphincterremoving surgeryVariable Received pRTnumber (% or  ± SD)Did not receive pRTnumber (% or  ± SD)OR (95%CI)  P  valueAge 62.2 ( ± 11.6) 67.9 ( ± 11.5) 0.959 (0.945–0.973)  \ 0.0001 a SexMale 137 (36) 245 (64)Female 99 (33) 201 (67) 1.135 (0.825–1.562) 0.4353GeographicNorth 111 (35) 209 (65) 0.810 (0.576–1.138) 0.2246 b LocationCenter 101 (40) 154 (60) 2.268 (1.350–3.811) 0.0020South 24 (22) 83 (78) 1.837 (1.104–3.057) 0.0222Center case \ 30 101 (31) 228 (69)Volume  C 30 135 (38) 218 (62) 0.715 (0.521–0.983) 0.0388Community hospital 139 (33) 288 (67)University hospital 97 (38) 158 (62) 0.786 (0.569–1.087) 0.1454APR d 49 (48) 54 (52)LAR c 176 (33) 360 (67) 1.856 (1.211–2.844) 0.0045Laparoscopic 38 (47) 66 (53)Open 189 (46) 353 (54) 1.075 (0.699–1.620) 0.7720No complication 188 (34) 357 (66)Complications 48 (35) 89 (65) 0.976 (0.659–1.446) 0.9052 Table 6  Multivariate logistic regression analysis with stepwise pro-cedure: factors associated with preoperative radiotherapy deliveryOdd ratio estimatesEffect Characteristic Estimate (95% CI)  P  valueGeographiclocationNorth vs South 1.823 (1.069–3.109) 0.0275Center vs South 2.836 (1.627–4.942) 0.0002Center case volume Small vs large 0.660 (0.465–0.938) 0.0203Age 1 year increase 0.954 (0.940–0.969)  \ 0.0001Sphincter saving APR vs LAR 1.988 (1.263–3.135) 0.0030Tech Coloproctol (2010) 14:229–235 233  1 3
Similar documents
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks