Search

Background: We evaluated the value of pre-treatment positron-emission tomography–computed tomography (PET-CT)-based radiomic features in predicting the locoregional progression-free survival (LR-PFS) of patients with inoperable or unresectable oesophageal cancer. Material and Methods: Forty-six patients were included and 230 radiomic parameters were extracted. After a principal component analysis (PCA), we identified the more robust radiomic parameters, and we used them to develop a heatmap. Finally, we correlated these radiomic features with LR-PFS. Results: The median follow-up time was 17 months. The two-year LR-PFS and PFS rates were 35.9% (95% CI: 18.9–53.3) and 21.6% (95%CI: 10.0–36.2), respectively. After the correlation analysis, we identified 55 radiomic parameters that were included in the heatmap. According to the results of the hierarchical clustering, we identified two groups of patients presenting statistically different median LR-PFSs (22.8 months vs. 9.9 months; HR = 2.64; 95% CI 0.97–7.15; p = 0.0573). We also identified two radiomic features (“F_rlm_rl_entr_per” and “F_rlm_2_5D_rl_entr”) significantly associated with LR-PFS. Patients expressing a “F_rlm_2_5D_rl_entr” of

Routine Electronic Monitoring of Health-Related Quality of Life (HRQoL) (REMOQOL) in clinical care with real-time feedback to physicians could help to enhance patient-centered care. We evaluated the feasibility of REMOQOL in the French context in the QOLIBRY study. The primary objective was to assess the patients’ compliance with REMOQOL. The QOLIBRY study was a single-center, prospective study conducted in the University Hospital of Besancon (France). Eligible patients were those treated with systemic therapies for breast, lung or colorectal cancer at any stage. Patients were invited to complete the EORTC QLQ-C30 questionnaire and cancer-site-specific modules before each visit on tablets and/or computers in the hospital or at home. During the consultation, physicians had real-time access to visual summaries of HRQoL scores. Compliance was assessed as adequate if at least 66% of HRQoL assessments were completed during the 4 months of follow-up. Between March 2016 and October 2018, 177 patients were included in the QOLIBRY study. Median age was 64 years (IQR 54–71). The proportion of patients with an adequate compliance rate was 95.5% (n = 63) in the breast cancer cohort, 98.2% (n = 55) in the colorectal cancer cohort, and 90.9% (n = 50) in the lung cancer cohort. The physicians checked the HRQoL results in 73.1% of visits and prescribed supportive care and adapted patient management in 8.3% and 5.2% of visits, respectively. The results of QOLIBRY study suggest that REMOQOL is feasible in the French context. However, information about HRQoL monitoring, training of the physicians in the use of the software, and recommendations for using HRQoL results to guide care are essential and must be improved.

618 Background: Muscle metastases (MM) are among rare secondary locations in renal cell carcinoma (RCC). Current guidelines do not provide specific advice on the management of these patients. There is a lack of scientific literature on the subject, mainly based on case reports or small retrospective monocentric cohorts. To date, therefore, there remains uncertainty about the clinical history, prognosis, and appropriate management of patients with MM. Methods: ARTEMIS is an ambispective national French multicenter, non-interventional study. It was opened to patients with metastatic RCC who had MM. The study was designed to assess overall survival (OS), progression-free survival (PFS), describe diagnostic and treatment modalities, and treatment-related serious adverse events in case of local treatments (TR-SAEs). Results: Median follow-up was 74.4 months IC95% [38.7-84.1]. The 146 enrolled patients were 73.6% male, with a median age of 57.6 years at diagnosis. They were initially diagnosed with stage IV RCC for 40.4% of them and had a favorable (36.2%), intermediate (45.7%), or poor (18.1%) IMDC risk score at the onset of metastases. Initially, metastases were mainly located in the lungs (55.5%), lymph nodes (26.7%), and bones (24.0%). Patients had a history of partial or total nephrectomy in 78.8% of cases. There were 30.0% of patients with synchronous MM. The median times from initial diagnosis or metastatic status to MM discovery were 25.5 [16.8-35.7] months and 8.4 [5.0-13.5] months, respectively. The majority (69.0%) of MM were discovered on conventional injected CT scans, whereas 8.3% were diagnosed clinically. Survival data are displayed on the table. Thirty-seven (25.3%) patients received local treatment of their MM which consisted of: external radiotherapy (48.6%), surgery (27.0%), cryotherapy (27.0%), stereotactic radiotherapy (5.4%), embolization (2.7%). Among them, local treatments prevented local relapse of MM in 28 patients (75.7%). Local TR-SAEs were seen in two patients (5.4%). Conclusions: ARTEMIS is the largest published cohort describing the disease history of patients with RCC suffering from MM. To our knowledge, this is the first study reporting the diagnostic and treatment modalities, also survival data. Despite its low incidence, the issue of these patients is not so rare in physicians’ practice. This work thus allows a greater understanding of clinical features and treatment of this pathology. [Table: see text]

700 Background: Pts with aPDAC often have general health impairment due to high symptom burden at diagnosis. We investigated prospectively the clinical benefit (CB) of an EISCP followed by CTx as early as suspicion of aPDAC in pts with ECOG performance status (PS)≥2. Methods: In this multicenter study, PS≥2 pts with pathologically confirmed or suspected aPDAC on imaging were included at first oncology visit (V1) in a personalized 14-d EISCP including pain, nutritional, diagnostic/stenting procedures. Post-EISCP PS≤1 pts received mFOLFIRINOX or gemcitabine(Gem)/Nab-paclitaxel(NP), PS≥2 received mFOLFOX7 or investigator choice CTx or best supportive care (BSC). The primary endpoint was the 14-EISCP success on both feasibility of procedures within 14-d±2 and CB defined by post-EISCP PS≤1, ≥5 points improvement of either fatigue/pain/global health EORTC QLQ-C15-PAL quality of life (QOL) scores, or CTx start ≤30d, aiming 59% success for clinical relevance. Secondary endpoint included uni/multivariate median overall survival (mOS) analyses. Results: As of 07/2022, 106 pts were included; 93 pts were evaluable for primary endpoint: male 46%, mean age 74 yrs, PS2/3 79%/21%, metastases (M1) 62%, pathological diagnosis needed 53%, biliary stenting 16%. V1 mean QOL global health score was 48±23. mOS was 4.1 months (IC95 2.8-5.7). The 14-d feasibility was achieved in 71%. Post-EISCP CB was observed in 82% of pts: 13%, PS improvement to 0/1, 23% QOL improvement and/or 73% CTx start ≤30-d. Eight pts (9%) died during the EISCP. The primary endpoint was achieved in 59% of pts (n=55). Overall, 17 pts (16%) received mFOLFIRINOX/GemNP, 32 (30%) FOLFOX, 29 (27%) Gem or 5FU alone, and 28 (27%) BSC. OS analyses are shown. At d30, pts receiving CTx had a mean change of QOL global health score from 51±23 to 60±23. Conclusions: In PS≥2 pts with aPDAC, a personalized 14-EISCP is feasible and lead to a meaningful CB allowing the administration of doublet/triplet CTx in nearly half of the pts. Pts starting CTx within 30d had improved OS. Clinical trial information: NCT02979483 . [Table: see text]

A systematic literature review of immuno-oncology trials was conducted to assess the potential impact of open-label vs double-blind trial design on patient-reported outcome (PRO) data. A systematic search of indexed literature published from January 2009 to May 2019 was conducted using PubMed/MEDLINE, Cochrane Library, and EMBASE database. All randomized clinical trials (RCTs) of immuno-oncology therapies on advanced cancer patients reporting PRO data were identified. Descriptive analyses were performed to quantify differences at baseline and over time, by the type of study, regarding questionnaire completion rate and PRO scores. In total, 23 studies were retained (15 open-label, 8 blinded). At baseline, no difference in completion rate was observed between arms irrespective of trial design (absolute mean difference of 2.8% and 2.2% for open label and blinded studies, respectively). No clinically significant difference in baseline PRO scores was observed between arms. Over time, impact on PRO scores could not be identified due to the limited number of studies, heterogeneity of questionnaires and tumor types. Trial design had no impact on PRO completion rate or baseline scores. Future research should involve analyses by specific cancer types and ideally compare individual data from two similar RCTs (blinded vs. open-label).

4108 Background: D (anti-PDL1) plus T (anti-CTLA-4) combination immunotherapy showed encouraging results in hepato-biliary cancers. Its efficacy in non-Asian Pts with pretreated BTC is unknown. Methods: IMMUNOBIL GERCOR-D18-1 PRODIGE-57 was initially a 2-arm, open-label, randomized non-comparative phase II study. Pts with recurrent/advanced pathologically proven BTC (intrahepatic cholangiocarcinoma [iCCA]/extrahepatic CCA [eCCA]/gallbladder cancer [GC]), ECOG PS 0-1, pre-treated with platinum-based CTx were randomized (1:1) to D (1500 mg Q4W) plus T (75 mg Q4W x 4 cycles [T75]) (Arm A) or D plus T in combination with weekly paclitaxel (Arm B). Arm B was closed prematurely for toxicity after inclusion of 10 Pts. The study continued with Arm A only. It was further amended to modify the T schedule (300 mg at cycle 1 [T300], amended Arm A) due to higher efficacy reported in other tumors as compared to T75 mg x 4. The new primary endpoint was the overall survival (OS) rate at 6 months (M6) in amended Arm A (D + T300) with a Fleming two-stage design (H0: 50%, H1: 65%, one-sided alpha: 5%, power: 90%). A total of 100 evaluable Pts were required (efficacy threshold: 59%). We present here the efficacy data of Arm A (D + T75). Results: From 12/2018 to 12/2020, 106 Pts were included in Arm A; 103 were evaluable for OS at M6. Median age was 66 years, 47% were male, 46% had ECOG PS 0, 69%/18%/13% had iCCA/eCCA/GC, 76% had metastatic disease, and 28% had prior tumor resection. First-line CTx was GEMCIS/GEMOX/5-FU-based/other in 63%/22%/4%/11%. The M6-OS rate was 59.2%. With a median follow-up of 12 months (95% Cl 11.4-14.9), the median OS was 8.0 months (95% Cl 5.7-11.7) and median PFS was 2.5 months (95% Cl 2.0-3.2). Complete response (CR) was observed in 2 (1.9%) Pts, partial response (PR) in 8 (7.8%), and stable disease (SD) in 32 (31.1%), resulting in an objective response rate of 9.7% and a disease control rate of 40.8%. Absence of progression (PD, iRECIST) after 2 cycles (M6-OS rate: 84% vs 41%, median OS: 17.9 months vs 4.4 months) and CR/PR as a best overall response (M6-OS rate: 100% vs 84% and 39% for SD and PD, respectively) were associated with markedly prolonged OS. 65 (63.1%) Pts had ≥1 grade 3-4 (G3-4) adverse event (AE) and 22 (21.4%) had ≥1 G3-4 treatment-related AE (TRAE). The most commonly reported G3-4 AEs were fatigue (12.6%), abdominal pain (5.8%), and aspartate aminotransferase increase (5.8%) and G3-4 TRAEs were fatigue (4.9%) and diarrhea (2.9%). One death was possibly related to treatment. Conclusions: Although no statistical conclusions can be drawn from this exploratory analysis of Arm A, D+T75 reached the pre-defined threshold for efficacy, with no unexpected toxicity. Results from amended Arm A (D+T300, N = 106 additional Pts), quality of life, and ancillary studies are pending. Clinical trial information: NCT03704480.

Background Randomized controlled trials (RCTs) represent the best evidence in oncology research. Glioblastoma is the most frequent and deadly primary brain tumor, affecting health-related quality of life. An important end point is patient-reported outcomes (PROs). There are no data regarding how well publications of glioblastoma RCTs report PROs. A specific PRO extension of the Consolidated Standards of Reporting Trials (CONSORT) statement was created to improve the quality of reporting. The aim of this study was to evaluate adherence to the CONSORT-PRO statement in reporting RCTs addressing the treatment of patients with glioblastoma. PRO analysis methodology was explored and criteria associated with higher quality of reporting were investigated. Methods From PubMed/MEDLINE and the Cochrane Library databases, all phase 2 and 3 RCTs related to glioblastoma published between 1995 and 2018 were reviewed according to the CONSORT-PRO statements. An overall quality score on a 0 to 100 scale was defined based on these criteria and factors associated with this score were identified. Results Forty-four RCTs were identified as relevant according to predefined criteria. The median overall quality score was 26. No difference was observed regarding reporting quality over the years. CONSORT-PRO items concerning data collection and analysis were poorly reported. Thirty-four trials (77%) used longitudinal data. The most frequent statistical method for PROs analysis was the mean change from baseline (63%). Factors associated with improved overall quality score were the presence of a secondary publication dedicated to PROs results, the statement of any targeted dimensions, and when trials reported results using multiple methods. Conclusion Despite the importance of measuring PROs in patients with glioblastoma, employment of the CONSORT-PRO statement is poor in RCTs.

Background and objective : Longitudinal analysis of patient-reported outcome (PRO) data remains challenging, as no standardization of statistical methods has been proposed, making comparison of PRO results between clinical trials difficult. In this context, the time to deterioration approach has recently been proposed and is regularly used as a modality of longitudinal PRO analysis in oncology. Methods : Two new SAS macro programs were developed, %TTD and %TUDD , which implement longitudinal analysis of PRO data according to the time to deterioration approach. These programs implement the recommended deterioration definitions. We described the programs with their different functionalities. Results : The %TTD macro calculates the time to first or transient deterioration, and the %TUDD macro calculates the time until definitive deterioration. These macros allow to obtain the survival variables from the time to deterioration approach. We illustrate our programs by presenting different applications on the randomized phase II AFUGEM GERCOR clinical trial. Conclusion : The implementation of the deterioration definitions in SAS software allows the dissemination of this approach, in order to move toward the goal of standardization of longitudinal PRO analysis in oncology clinical trials.

Introduction L’e-monitoring de la qualite de vie relative a la sante (QdV) en pratique clinique courante (« Routine Electronic Monitoring of Health-Related Quality of Life [REMOQOL] ») consiste au recueil d’informations rapportees par le patient (symptomes et dimensions fonctionnelles) et a la transmission de ces informations en temps reel aux equipes soignantes. L’objectif est de fournir un outil complementaire pour la prise en charge des patients. L’etude QOLIBRY vise a evaluer la faisabilite de cette demarche sur une cohorte francaise de patients atteints de cancers. Methodes QOLIBRY est une etude prospective monocentrique realisee au CHU de Besancon. Les patients eligibles (traites pour un cancer du sein, du colon-rectum ou du poumon) ont ete invites a completer le questionnaire QLQ-C30 de l’EORTC ainsi que le module specifique associe a leur localisation respective, QLQ-BR23, QLQ-CR29 ou QLQ-LC13. Le logiciel CHES est utilise pour le remplissage des questionnaires de QdV par les patients avant chaque consultation. Les questionnaires sont completes a l’hopital ou depuis le domicile du patient, et les scores de QdV sont presentes de maniere graphique et instantanee aux cliniciens. L’objectif principal est l’evaluation du taux de compliance, defini comme le nombre de patients ayant repondu au minimum a 66 % des temps de mesure prevus sur le nombre de patients total. L’acceptabilite par les patients, l’exhaustivite, la faisabilite organisationnelle et informatique, l’utilisation des resultats par les cliniciens et leur satisfaction, ainsi que la maitrise de l’outil informatique par les patients sont les autres criteres de jugement. Resultats Entre mars 2016 et octobre 2018, 177 patients traites majoritairement par chimiotherapie et therapie ciblee ont ete inclus (sein n = 66, colon-rectum n = 56, poumon n = 55). Parmi les patients auxquels l’etude a ete proposee, 23 ont refuse de participer : 12 % (n = 11) dans la cohorte sein, 7 % (n = 5) dans la cohorte colon-rectum et 10 % (n = 7) dans la cohorte poumon. Plus de 1400 temps de mesure de QdV ont ete recueillis pendant les quatre mois de suivi des patients avec une duree mediane de sept jours entre chaque mesure pour la cohorte sein et de 14 jours pour les deux autres cohortes ; 36,7 % des questionnaires ont ete completes par les patients directement depuis leur domicile. Le taux de compliance est de 94,9 % pour l’ensemble de la population avec des resultats similaires pour chacune des cohortes, 95,5 % (n = 63) pour la cohorte sein, 98,2 % (n = 55) pour la cohorte colon-rectum et 90,9 % (n = 50) pour la cohorte poumon. Pour 73,1 % des temps de mesure, les medecins declarent avoir consulte les resultats de QdV. Suite a la consultation de ces resultats, des soins de support ont ete prescrits dans 8,3 % des cas et la strategie therapeutique a ete adaptee dans 5,2 % des cas. Conclusion Les resultats de QOLIBRY suggerent que REMOQOL est faisable sur une cohorte de patients francaise et bien accepte par les patients. La formation des medecins et les recommandations d’utilisation des resultats de la QdV pour guider les soins sont egalement des points cles et doivent etre ameliorees. Deux autres etudes sont en cours, QUANARIE, etude multicentrique evaluant la faisabilite de REMOQOL dans la prise en charge en routine de patients porteurs d’un adenocarcinome du rein initiant un traitement par ITK anti-VEGF oral, et l’etude randomisee REMOQOL-poumon qui permettra de comparer la qualite de la relation de soin percue avec et sans REMOQOL.

Introduction L’e-monitoring de la qualite de vie relative a la sante (QdV) des patients atteints de cancer (« Routine Electronic Monitoring of Health-Related Quality of Life » [REMOQOL]) est une demarche recente dans laquelle la QdV des patients est recueillie systematiquement et transmise en temps reel aux equipes medicales, dans le but de fournir un outil complementaire pour la prise en charge des patients. Cela peut notamment permettre l’orientation vers des soins de support personnalisee, l’evaluation systematique des toxicites et l’adaptation de la strategie therapeutique. Deux etudes, QOLIBRY et QUANARIE, ont ete mises en œuvre pour evaluer la faisabilite de REMOQOL. Methodes La satisfaction des investigateurs vis-a-vis de REMOQOL et des outils associes a ete evaluee dans deux etudes : QOLIBRY (patients atteints de cancer du sein, poumon ou colon-rectum) est une etude monocentrique et l’etude QUANARIE (rein metastatique) est multicentrique (neuf centres). Dans ces deux etudes, le patient est invite a repondre aux questionnaires de QdV de l’EORTC avant chaque consultation, en utilisant le logiciel CHES sur tablette et/ou ordinateur, a l’hopital ou depuis son domicile. Le CHES permet la presentation instantanee des resultats de QdV sous forme graphique. Avant ou pendant la visite, le medecin consulte les donnees de QdV et peut ainsi constater l’evolution de la QdV de son patient et adapter la prise en charge en consequence. Nous avons developpe un questionnaire compose de 43 items pour evaluer la satisfaction des medecins vis-a-vis de REMOQOL. Les items etaient sous forme de questions a choix ou d’echelles de Likert de 0 (pas du tout d’accord avec l’affirmation proposee) a 10 (tout a fait d’accord). Les reponses ont ensuite ete categorisees comme tel : 0–4 pas du tout d’accord, 4–6 neutre et 7–10 tout a fait d’accord. Le logiciel CleanWeb est utilise pour repondre a ce questionnaire et pour une compliance optimale des rappels automatiques ont ete programmes. Resultats Entre septembre 2016 et mars 2019, 50 investigateurs ont inclus 249 patients (QOLIBRY n = 193, QUANARIE n = 56) ; 31 (67,4 %) investigateurs ont complete l’enquete. Parmi eux, 22 (71 %) ont declare avoir consulte les resultats de la QdV, dont 15 (68,2 %) etaient d’accord avec l’affirmation « Les resultats de QdV aident a mieux comprendre l’etat de sante du patient ; 86,4 % (n = 19) etaient d’accord avec l’affirmation « Il est facile d’aborder les resultats de QdV avec le patient » et « L’utilisation des resultats de QdV ameliore les echanges entre le patient et le medecin » (54,5 %, n = 12). Treize medecins (59,1 %) etaient d’accord avec l’affirmation « L’interpretation des resultats de QdV aide a l’adaptation de la prise en charge du patient » et « L’utilisation des resultats de QdV aide lors de la prescription de soins de support » (54,5 %, n = 12) sans pour autant prolonger la duree de la consultation (50 %, n = 11). Cependant, 36,4 % (n = 8) des investigateurs ne consideraient pas que les resultats etaient facilement interpretables. Au final, La majorite des medecins interroges, 76,7 % (n = 23) seraient d’accord pour integrer REMOQOL dans leur pratique clinique courante pour tous les patients. Conclusion Les medecins ayant repondu a cette enquete ont ete globalement satisfaits et semblent prets a utiliser REMOQOL dans leur pratique clinique courante. Cependant, la formation des medecins a l’interpretation des donnees de QdV doit etre amelioree pour impliquer tous les medecins.

PURPOSE Tumor deposits (TDs) seem to affect the prognosis of patients with colon cancer (CC). In the seventh edition of the American Joint Committee on Cancer TNM staging system for CC, the presence of TDs is only considered in the absence of lymph node metastases (LNMs). In the era of personalized duration of histopathologic criteria-based adjuvant therapy, this could potentially lead to a decrease in the prognostic prediction accuracy. PATIENTS AND METHODS A post hoc analysis of all pathologic reports from patients with stage III CC included in the IDEA France phase III study (ClinicalTrials.gov identifier: NCT00958737 ) investigating the duration of adjuvant fluorouracil, leucovorin, and oxaliplatin or capecitabine and oxaliplatin therapy (3 v 6 months) was performed. The primary objective was to determine the prognostic impact of TD on disease-free survival (DFS). The effect of the addition of TD to LNM count on pN restaging was also evaluated. A multivariable analysis was performed to establish the association between TD and DFS. RESULTS Of 1,942 patients, 184 (9.5%) had TDs. The pN1a/b and pN1c populations showed similar DFS. TD-positive patients had worse prognosis compared with TD-negative patients, with 3-year DFS rates of 65.6% (95% CI, 58.0% to 72.1%) and 74.7% (95% CI, 72.6% to 76.7%; P = .0079), respectively. On multivariable analysis, TDs were associated with a higher risk of recurrence or death (hazard ratio [HR], 1.36; P = .0201). Other adverse factors included pT4 and/or pN2 disease (HR, 2.21; P < .001), the 3 months of adjuvant treatment (HR, 1.29; P = .0029), tumor obstruction (HR, 1.28; P = .0233), and male sex (HR, 1.24; P = .0151). Patients restaged as having pN2 disease (n = 35, 2.3%) had similar DFS as patients initially classified as pN2. CONCLUSION The presence of TDs is an independent prognostic factor for DFS in patients with stage III CC. The addition of TD to LNM may help to better define the duration of adjuvant therapy.

3519 Background: Tumor deposits (TDs) are isolated tumor foci in the pericolic, perirectal or mesocolic fat without residual lymph node (LN) tissue. TDs seem to impact the prognosis of stage III colon cancer (CC) patients (pts) but are only considered in TNM staging in the absence of LN metastases (LNM). We aimed at evaluating the prognosis value for disease free survival (DFS)of TDs in International Duration Evaluation of Adjuvant Chemotherapy (IDEA) France phase III study (NCT00958737) that compared 3 versus 6 months of adjuvant FOLFOX or CAPOX for stage III CC pts. Methods: All pathological reports of pts included in IDEA France trial were retrospectively analyzed. DFS according to the presence or absence of TDs was evaluated using Kaplan-Meier estimator. Multivariable Cox model analysis was performed to evaluate the association between TDs and DFS. This analysis did not included immunohistochemical biomarkers. Results: Among the 2022 pts included in IDEA France study, 1942 (96%) were analyzed. 80 pts were excluded: no pathological report (n = 68), pts without treatment (n = 12). TDs were found in 184 pts (9.47%), of whom 74 with N1a/b (40%), 55 with N1c (30%) and 55 with N2 LN stage (30%). All characteristics were similar according to the presence of TDs, except for tumor/node (TN) stage (T4 and/or N2 are more frequent in pts with TDs; p = .0046). The 3-year DFS rates were 65.59% [95% confidence interval (95%CI) 58.04-72.12] and 74.71% [95%CI 72.57-76.71] for pts with and without TDs, respectively (p = 0.0079). In multivariable analysis, TDs were associated with higher risk of recurrence or death (hazard ratio (HR) = 1.36, 95%CI 1.05-1.75, p = .0201), as well as T4 and/or N2 (HR = 2.21, 95%CI 1.03-1.59, p < .001), 3 months of adjuvant treatment (HR = 1.29, 95%CI 1.09-1.52, p = .0029), obstruction (HR = 1.28, 95%CI 1.03-1.59, p = .0233) and male (HR = 1.24, 95%CI 1.04-1.46, p = .0151). Adding TDs count to the LNM count, 35 out of 1454 N1a/b/c CC pts (2.4%) were reclassified as N2 and experienced worse 3 years DFS than confirmed N1 CC pts (p = .0151). Conclusions: TD is an independent and valuable prognostic factor for DFSin stage III CC pts and should be considered whatever the LNM status.

497 Background: Recent reports demonstrate prognostic and predictive impacts of the location of the primary tumor in metastatic colorectal cancer (mCRC). Our retrospective analysis aimed to determine the influence of primary site on metastatic distribution and disease evolution. Methods: From our database all patients (pts) with mCRC (except transverse carcinoma) treated from 1/12/2007 to 1/12/2016 in our institution were collected. Univariate and multivariate analyses were performed to identify predictors of overall survival (OS). Results: A total of 284 pts with available data were analyzed: 83 with Right-sided Colon Cancer (RCC) (29%), 123 with Left-sided Colon Cancer (LCC) (43%) and 78pts with Rectal Cancer (RC) (28%). Hepatic, lung, lymph nodes and peritoneal metastases were respectively found in 63%, 36%, 23% and 20% of the population. The incidence or number of liver metastases were not influenced by sidedness (p = 0.06), LCC presented more bilobar involvement compared to RCC and RC (p = 0.017). Peritoneal carcinomatosis was significantly correlated to colon cancer (p = 0.002), whereas lung metastases were more common in RC (p < 0.001). Patients with RCC more often presented distal lymph node involvement (p = 0.008). RAS mutation status was known for 241pts (80%), of those 110 (37%) were RAS mutated with no significant differences between RCC, LCC and RC (p = 0.4), BRAF mutation (p = 0.007) was more common in RCC. On a multivariable analysis, primary tumor resection (PTR) and complete response after first line therapy were associated with a better OS but only a trend was observed for LCC and RC. Lung, lymph and peritoneal metastasis were associated with worse OS (Table). Conclusions: These results suggests that mCRC had different clinical presentation at diagnosis, the association with molecular features may explain the independent prognostic factor for OS of the sidedness. [Table: see text]