Pascal Garaud

Final Results of the 94–01 French Head and Neck Oncology and Radiotherapy Group Randomized Trial Comparing Radiotherapy Alone With Concomitant Radiochemotherapy in Advanced-Stage Oropharynx Carcinoma

PURPOSE We report the 5-year survival and late toxicity results of a randomized clinical trial, which showed a 3-year improvement in overall survival and locoregional control of stage III or IV oropharynx carcinoma, using concomitant radiochemotherapy (arm B), compared with standard radiotherapy (arm A). PATIENTS AND METHODS A total of 226 patients were entered onto a phase III multicenter randomized trial comparing radiotherapy alone (70 Gy in 35 fractions; arm A) with concomitant radiochemotherapy (70 Gy in 35 fractions with three cycles of a 4-day regimen comprising carboplatin and fluorouracil; arm B). Prognostic factors were evaluated by univariate and multivariate analysis. Five-year late toxicity was evaluated using National Cancer Institute Common Toxicity Criteria for neurological toxicity, hearing, taste, mandibula, and teeth damage, and Radiation Therapy Oncology Group toxicity criteria for skin, salivary gland, and mucosa. RESULTS Five-year overall survival, specific disease-free survival, and locoregional control rates were 22% and 16% (log-rank P =.05), 27% and 15% (P =.01), and 48% and 25% (P =.002), in arm B and arm A, respectively. Stage IV, hemoglobin level lower than 125 g/L, and standard treatment were independent prognostic factors of short survival and locoregional failure by univariate and multivariate analysis. One or more grade 3 to 4 complications occurred in 56% of the patients in arm B, compared with 30% in arm A (P was not significant). CONCLUSION Concomitant radiochemotherapy improved overall survival and locoregional control rates and does not statistically increase severe late morbidity. Anemia was the most important prognostic factor for survival in both arms.

Randomized Trial of Induction Chemotherapy With Cisplatin and 5-Fluorouracil With or Without Docetaxel for Larynx Preservation

BACKGROUND Chemotherapy with cisplatin (P) and 5-fluorouracil (F) followed by radiotherapy in patients who respond to chemotherapy is an alternative to total laryngectomy for patients with locally advanced larynx and hypopharynx cancer. Data suggest that docetaxel (T) may add to the efficacy of PF. The objective of this trial was to determine whether adding T to PF could increase the larynx preservation rate. METHODS Patients who had larynx and hypopharynx cancer that required total laryngectomy were randomly assigned to receive three cycles of TPF or PF. Patients who responded to chemotherapy received radiotherapy with or without additional chemotherapy. Patients who did not respond to chemotherapy underwent total laryngectomy followed by radiotherapy with or without additional chemotherapy. The primary endpoint was 3-year larynx preservation rate. Secondary endpoints included acute toxicities and overall response. All statistical tests were two-sided. RESULTS Baseline patient and tumor characteristics were well balanced between the TPF (n = 110) and PF (n = 103) groups. With a median follow-up of 36 months, the 3-year actuarial larynx preservation rate was 70.3% with TPF vs 57.5% with PF (difference = 12.8%; P = .03). Patients in the TPF group had more grade 2 alopecia, grade 4 neutropenia, and febrile neutropenia, whereas patients in the PF group had more grade 3 and 4 stomatitis, thrombocytopenia, and grade 4 creatinine elevation. The overall response was 80.0% in the TPF group vs 59.2% in the PF group (difference = 20.8%; P = .002). CONCLUSIONS In patients with advanced larynx and hypopharynx carcinomas, TPF induction chemotherapy was superior to the PF regimen in terms of overall response rate. These results suggest that larynx preservation could be achieved for a higher proportion of patients.

Scapular notching in reverse shoulder arthroplasty

The causes and consequences of scapular notching after reverse shoulder arthroplasty (RSA) were investigated in 326 consecutive patients (337 shoulders) undergoing RSA between 1991 and 2003. Patients underwent 269 (80%) primary RSAs and 68 revisions of unconstrained shoulder prosthesis. At last follow-up (average, 47 months; range, 24-120 months) 62% had scapular notching. Notching frequency and extension were correlated to the length of follow-up (P = .0005). Notching was more frequent in cuff tear arthropathy (P = .0004), grade 3 or 4 fatty infiltration of the infraspinatus (P = .01), and narrowed acromiohumeral distance (P < .0001). Glenoids preoperatively oriented superiorly were more at risk for notching (P = .006). More notching occurred when the RSA was implanted using an anterosuperior approach vs a deltopectoral approach (P < .0001). Notching was correlated with humeral radiolucencies in proximal zones (P < .0001) and with glenoid radiolucent lines (P < .0001). Positioning of the baseplate definitely influences scapular notching. High positioning of the baseplate and superior tilting must be avoided.

Induction Chemotherapy Followed by Either Chemoradiotherapy or Bioradiotherapy for Larynx Preservation: The TREMPLIN Randomized Phase II Study

PURPOSE To compare the efficacy and safety of induction chemotherapy (ICT) followed by chemoradiotherapy (CRT) or bioradiotherapy (BRT) for larynx preservation (LP). PATIENTS AND METHODS Previously untreated patients with stage III to IV larynx/hypopharynx squamous cell carcinoma received three cycles of ICT-docetaxel and cisplatin 75 mg/m(2) each on day 1 and fluorouracil 750 mg/m(2) per day on days 1 through 5. Poor responders (< 50% response) underwent salvage surgery. Responders (≥ 50% response) were randomly assigned to conventional radiotherapy (RT; 70 Gy) with concurrent cisplatin 100 mg/m(2) per day on days 1, 22, and 43 of RT (arm A) or concurrent cetuximab 400 mg/m(2) loading dose and 250 mg/m(2) per week during RT (arm B). Primary end point was LP at 3 months. Secondary end points were larynx function preservation (LFP) and overall survival (OS) at 18 months. RESULTS Of the 153 enrolled patients, 116 were randomly assigned after ICT (60, arm A; 56, arm B). Overall toxicity of both CRT and BRT was substantial following ICT. However, treatment compliance was higher in the BRT arm. In an intent-to-treat analysis, there was no significant difference in LP at 3 months between arms A and B (95% and 93%, respectively), LFP (87% and 82%, respectively), and OS at 18 months (92% and 89%, respectively). There were fewer local treatment failures in arm A than in arm B; salvage surgery was feasible in arm B only. CONCLUSION There is no evidence that one treatment was superior to the other or could improve the outcome reported with ICT followed by RT alone (French Groupe Oncologie Radiothérapie Tête et Cou [GORTEC] 2000-01 trial [Induction CT by Cisplatin, 5FU With or Without Docetaxel in Patients With T3 and T4 Larynx and Hypopharynx Carcinoma]). The protocol that can best compare with RT alone after ICT is still to be determined.

Late toxicity results of the gortec 94-01 randomized trial comparing radiotherapy with concomitant radiochemotherapy for advanced-stage oropharynx carcinoma: Comparison of LENT/SOMA, RTOG/EORTC, and NCI-CTC scoring systems

PURPOSE To prospectively assess 5-year late toxicity in patients treated by concomitant radiochemotherapy for locally advanced oropharynx carcinoma using three different toxicity scales. METHODS AND MATERIALS A total of 226 patients were entered in a Phase III multicenter, randomized trial comparing radiotherapy alone (70 Gy in 35 fractions: Arm A) with concomitant radiochemotherapy (70 Gy in 35 fractions with three cycles of a 4-day regimen containing carboplatin and 5-fluorouracil: Arm B). Five living patients, free of local or distant recurrences, could not be evaluated for late toxicity. Forty-four patients were eligible for late toxicity with a median follow-up of 5 years. Late toxicity was evaluated by the radiation oncologist using a large questionnaire containing 120 mixed items of three scales (NCI-CTC, LENT/SOMA, and RTOG). The data were then transposed on separate scales using corresponding grades. RESULTS The 5-year overall survival rate was 22% in Arm B and 16% in Arm A (p = 0.05). The 5-year locoregional control rate was 48% in Arm B and 25% in Arm A (p = 0.002). The spinal cord was not affected by the concomitant adjunct of chemotherapy, and no deaths were caused by late toxicity. Using the three late toxicity scales, 100% of the patients treated with the combined modality (Arm B) developed one or more late complications vs. 94% in the radiotherapy-alone arm (Arm A). The difference was not statistically significant. The most commonly damaged organs (all Grade 1-4) were the salivary glands (100% in Arm B vs. 82% in Arm A, p <0.05), skin (78% vs. 47%, p <0.05), teeth (67% vs. 18%, p <0.05), mucosa (59% vs. 63% p = not significant), and mandible (44% vs. 12%, p <0.05). One or more Grade 3-4 complications occurred in 82% of the patients in Arm B vs. 47% in Arm A (p = 0.02) but concerned only the teeth. The correlation between the RTOG and LENT/SOMA scale and between the NCI-CTC and LENT/SOMA scale were low for Grade 1-4 toxicity (near 30%). The transposability of a patients symptoms was significantly greater using the LENT/SOMA or RTOG/EORTC scaling systems than using the NCI-CTC system. CONCLUSION Concomitant radiochemotherapy increased overall survival and locoregional control rates. The difference between the two treatment groups for Grade 3-4 complications was only significant for the teeth. The late toxicity assessment of a treatment may depend on the toxicity scale used. The LENT/SOMA scale seems to be the most accurate scale, but most of the score results were not concordant with those obtained with other scales. The results of this study confirm the necessity of using a common late toxicity scale in clinical trials.

Preoperative radiotherapy (RT) for rectal cancer: Predictive factors of tumor downstaging and residual tumor cell density (RTCD): Prognostic implications

PURPOSE To determine predictive factors and prognostic value of tumor downstaging and tumor sterilization after preoperative RT for rectal cancer. METHODS AND MATERIALS Between 1977 and 1994, 167 patients with a histologically proven adenocarcinoma (70 T2, 65 T3, 29 T4, and 3 local recurrences) underwent preoperative RT. Median dose was 44 Gy (5-73 Gy). Surgery was performed in a mean time of 5 weeks after RT. Pathologic specimens have been reviewed by the same pathologist in order to specify the modified Astler Coller classification (MAC), and to quantify the residual tumor cell density (RTCD). RESULTS According to the MAC, there was 9 stage 0 (5%), 10 stage A (6%), 103 stage B1-B3 (62%), and 45 stage C1-C3 (27%) tumors. Seventeen percent and 56% of the patients who received a dose > or = 44 Gy had respectively a 0-A and a B tumor, compared to 4 and 69% in those who received a dose < 44 Gy (p = 0.04). Tumor differentiation and a longer interval before surgery were significantly associated with a more frequent downstaging, and preoperative staging correlated well to the postoperative pathological findings. According to the RTCD, 62 tumors (37%) showed no or only rare foci of residual tumor cells (Group 1); 62 (37%) showed an intermediate RTCD (Group 2); and 43 (26%) a high RTCD (Group 3). No predictive factor of RTCD was statistically significant. In univariate analysis, postoperative staging was a significant prognostic factor, with corresponding 5-year overall survival rates in 0-A, B, and C stages of 92, 67, and 26% (p < 0.01). RTCD was not a prognostic factor. However, overall and disease-free survival rates for patients with complete pathologic response of 83% at 2 and 5 years suggested a better outcome in this subgroup of patients. CONCLUSION The favorable influence of higher doses of preoperative RT on pathologic stage has been observed. Tumor differentiation, preoperative classification and time before surgery were the other predictive factors of tumor downstaging. However, there was no predictive factor of complete pathologic response. Even after preoperative RT, postoperative staging remained a prognostic factor.

RADIOTHERAPY WITH CONCOMITANT WEEKLY DOCETAXEL FOR STAGES III/IV OROPHARYNX CARCINOMA. RESULTS OF THE 98-02 GORTEC PHASE II TRIAL

PURPOSE We designed a prospective Phase II clinical trial to evaluate the addition of weekly chemotherapy using Docetaxel during standard radiation therapy in patients with Stages III and IV oropharynx carcinoma. METHODS A total of 63 patients have been entered in a Phase II multicenter trial. Radiotherapy delivered, with conventional fractionation, 70 Gy in 35 fractions. Patients received during the period of radiotherapy seven cycles of Docetaxel (20 mg/m2 each week). RESULTS Radiotherapy compliance was good in respect to total dose, treatment duration, and treatment interruption. The rate of Grade 3 and 4 mucositis was 84%. Grade 3 and 4 skin toxicity occurred in 53% of the patients. Hematologic toxicity was infrequent, with only a 5% rate of Grade 3 or 4 neutropenia. Three-year overall actuarial survival and disease-free survival rates were, respectively, 47% (95% CI = 39-68%) and 39% (95% CI = 30-57%). The local and regional control rate was 64%. CONCLUSION The adjunction of weekly Docetaxel to conventional radiotherapy is feasible. Mucositis and skin toxicity were the major acute toxic effects. Therapeutic results were similar to those observed with concomitant chemotherapy using platinum and/or 5-FU. Further trials using Docetaxel in combination with other cytotoxic agents are needed.

Phase III Trial of Concurrent or Sequential Adjuvant Chemoradiotherapy After Conservative Surgery for Early-Stage Breast Cancer: Final Results of the ARCOSEIN Trial

PURPOSE In 1996, we initiated the French multicenter phase III randomized trial to compare the effect on disease-free survival (DFS) of concurrent versus sequential chemotherapy (CT) and radiotherapy (RT) after breast-conserving surgery for stages I and II breast cancer. This report presents the clinical results with a median follow-up of 60 months. PATIENTS AND METHODS Between February 1996 and April 2000, 716 patients were entered onto this trial. Adjuvant treatment began within 6 weeks after surgery. Sequential treatment of CT administered first followed by RT was compared with concurrent treatment of CT administered with RT. The CT regimen consisted of mitoxantrone (12 mg/m2), fluorouracil (500 mg/m2), and cyclophosphamide (500 mg/m2) on day 1, and it was repeated every 21 days for six courses. RT was delivered to the breast and, when indicated, to the regional lymphatics. RESULTS There was no statistically significant difference in treatment in the 5-year DFS (80% in both groups; P = .83), locoregional recurrence-free survival (LRFS; 92% in sequential v 95% in concurrent; P = .76), metastasis-free survival (87% in sequential v 84% in concurrent; P = .55), or overall survival (90% in sequential v 91% in concurrent; P = .76). Nevertheless, in the node-positive subgroup, the 5-year LRFS was statistically better in the concurrent arm (97% in concurrent v 91% in sequential; P = .02), corresponding to a risk of locoregional recurrence decreased by 39% (hazard ratio, 0.61; 95% CI, 0.38 to 0.93). CONCLUSION This treatment protocol remains an appealing clinical option for many women with operable breast cancer at a high risk of recurrence. Combination treatments with new drugs for breast cancer are warranted.

Omarthrose ecentrée: symposium

Cuff tear arthropathy is defined as the combination of a gleno-humeral arthritis and a massive rotators cuff tear. It is generally admitted that the cuff tear, or its deficiency jeopardises the results of anatomic prosthesis. Grammont imagined and grew the concept of the reverse prosthesis whose aim was to remedy the insufficiency of the rotator cuff and whose use has dramatically modified the therapeutic approach in these complicated situations. The aim of this symposium is to analyse the results of shoulder arthroplasty (anatomic prosthesis, bipolar or reverse) in cuff tear arthropathy, in massive and isolated cuff tears which justify the use of prosthetic surgery, and in centered osteoarthritis with deficient cuff also justifying this use. Massive cuff tears and cuff tear arthropathy have been considered as the stages of a same pathology by Hamada whose classification has been used for the purposes of this study. Out of the 738 initial prosthesis, 111 have been eliminated because of death, incomplete files or lost of sight, without any known complication. Out of the remaining 627 who were used as a basis for this symposium (representing 85% of initial cases), 570 who still had their prosthesis were reviewed and their functional results analysed after 2 years or more. The population was mainly female (72%) who were injured on the dominant side (75%) with a mean age of 72 years. The preoperative Constant score was in average 24 and 24% of the patients had already been operated on their shoulder. For those who had an acetabulization of the acromion, the strength in external rotation was significantly less satisfying and the lesion of the infraspinatus and the teres minor were more frequent. 48 hemiarthroplasties, 52 bipolar and 527 reverse prosthesis were studied. At revision, with an average follow up of 52 months, the revision rate was 23% for hemiarthroplasties, 14% for reversed prosthesis and only 8% for bipolar prosthesis. The prosthesis was removed in 21% of hemiarthroplasties, 5% of reverse and 2% of bipolar. No infection to report in the hemi group or the bipolar group, whereas there was an infection rate of 5% in the reverse group. Nevertheless, the Constant score was significantly better with reverse (62) than with bipolar (45) or hemi (44). The active elevation was also better with the reverse whereas the external rotation was not as good as with anatomic prosthesis. The analytic study of the results of the reverse prosthesis shows a negative influence of the lesion of the sub-scapularis and the teres minor. The results are disappointing with young patients and those who had surgical precedents. On X-ray, we can notice 0.5% of humeral loosening withouth any correlation to the fact that the implant is cemented or not, 3.6% of glenoid loosening and 68% of scapular notches without any significant change on the Constant score. Their occurrence is correlated to a preoperative rising of the humeral head and a superior glenoid lesion. It is observed more frequently with the supero-lateral approach compared to the delto-pectoral one. The frequency of these notches grows with the follow up and their occurrence is often associated to humeral radiolucent lines. On the long term, the survival rate of these prosthesis is 89% at 10 years. The Constant score deteriorates gradually after 7 years; this seems to be linked to the occurrence of x-ray modifications in the years that followed. The main complications observed with reverse prosthesis were the infections (5.1%), the glenoid problems (5.1%), the instabilities (3.6%), the acromion fractures (3.0%). Infections can be treated by a wash out and antibiotics in the first 3 months, then the removal of the prosthesis becomes necessary. The instabilities occur more often for males, with delto-pectoral approach and with 36mm diameter glenoids. The glenoid problems are frequent in the first years and often due to technical errors or material defects (unscrewing of the glenosphere). Acromion fractures have an important clinical impact when they concern the spine and there healing is difficult to obtain, whichever method is used. In conclusion, the use of a prosthesis for cuff tear arthropathies must be thought about, especially in massive cuff tear without osteoarthritis, in patients with previous surgery, and in patients younger than 70. If the active elevation is conserved and the patient is young, the use of an hemi or a bipolar prosthesis can be debated. In other cases, the indication of a reverse prosthesis is preferable given that the clinical results are better. In these cases, the surgical technique must be accurate, bearing in mind the advantages and disadvantages of the two possible approaches, the type of implant (36 vs 42), the position and orientation of the glenoid baseplate according to the pre-operative bone wear, the orientation of the humeral implant, the need for reinsertion of the subscapularis and, maybe, the possibility of an associated transfer of the latissimus dorsi.Cuff tear arthropathy is defined as the combination of a gleno-humeral arthritis and a massive rotators cuff tear. It is generally admitted that the cuff tear, or its deficiency jeopardises the results of anatomic prosthesis. Grammont imagined and grew the concept of the reverse prosthesis whose aim was to remedy the insufficiency of the rotator cuff and whose use has dramatically modified the therapeutic approach in these complicated situations. The aim of this symposium is to analyse the results of shoulder arthroplasty (anatomic prosthesis, bipolar or reverse) in cuff tear arthropathy, in massive and isolated cuff tears which justify the use of prosthetic surgery, and in centered osteoarthritis with deficient cuff also justifying this use. Massive cuff tears and cuff tear arthropathy have been considered as the stages of a same pathology by Hamada whose classification has been used for the purposes of this study. Out of the 738 initial prosthesis, 111 have been eliminated because of death, incomplete files or lost of sight, without any known complication. Out of the remaining 627 who were used as a basis for this symposium (representing 85% of initial cases), 570 who still had their prosthesis were reviewed and their functional results analysed after 2 years or more. The population was mainly female (72%) who were injured on the dominant side (75%) with a mean age of 72 years. The preoperative Constant score was in average 24 and 24% of the patients had already been operated on their shoulder. For those who had an acetabulization of the acromion, the strength in external rotation was significantly less satisfying and the lesion of the infraspinatus and the teres minor were more frequent. 48 hemiarthroplasties, 52 bipolar and 527 reverse prosthesis were studied. At revision, with an average follow up of 52 months, the revision rate was 23% for hemiarthroplasties, 14% for reversed prosthesis and only 8% for bipolar prosthesis. The prosthesis was removed in 21% of hemiarthroplasties, 5% of reverse and 2% of bipolar. No infection to report in the hemi group or the bipolar group, whereas there was an infection rate of 5% in the reverse group. Nevertheless, the Constant score was significantly better with reverse (62) than with bipolar (45) or hemi (44). The active elevation was also better with the reverse whereas the external rotation was not as good as with anatomic prosthesis. The analytic study of the results of the reverse prosthesis shows a negative influence of the lesion of the sub-scapularis and the teres minor. The results are disappointing with young patients and those who had surgical precedents. On X-ray, we can notice 0.5% of humeral loosening withouth any correlation to the fact that the implant is cemented or not, 3.6% of glenoid loosening and 68% of scapular notches without any significant change on the Constant score. Their occurrence is correlated to a preoperative rising of the humeral head and a superior glenoid lesion. It is observed more frequently with the supero-lateral approach compared to the delto-pectoral one. The frequency of these notches grows with the follow up and their occurrence is often associated to humeral radiolucent lines. On the long term, the survival rate of these prosthesis is 89% at 10 years. The Constant score deteriorates gradually after 7 years; this seems to be linked to the occurrence of x-ray modifications in the years that followed. The main complications observed with reverse prosthesis were the infections (5.1%), the glenoid problems (5.1%), the instabilities (3.6%), the acromion fractures (3.0%). Infections can be treated by a wash out and antibiotics in the first 3 months, then the removal of the prosthesis becomes necessary. The instabilities occur more often for males, with delto-pectoral approach and with 36mm diameter glenoids. The glenoid problems are frequent in the first years and often due to technical errors or material defects (unscrewing of the glenosphere). Acromion fractures have an important clinical impact when they concern the spine and there healing is difficult to obtain, whichever method is used. In conclusion, the use of a prosthesis for cuff tear arthropathies must be thought about, especially in massive cuff tear without osteoarthritis, in patients with previous surgery, and in patients younger than 70. If the active elevation is conserved and the patient is young, the use of an hemi or a bipolar prosthesis can be debated. In other cases, the indication of a reverse prosthesis is preferable given that the clinical results are better. In these cases, the surgical technique must be accurate, bearing in mind the advantages and disadvantages of the two possible approaches, the type of implant (36 vs 42), the position and orientation of the glenoid baseplate according to the pre-operative bone wear, the orientation of the humeral implant, the need for reinsertion of the subscapularis and, maybe, the possibility of an associated transfer of the latissimus dorsi.

Evaluation of the role of glenosphere design and humeral component retroversion in avoiding scapular notching during reverse shoulder arthroplasty

BACKGROUND Scapular notching is a common observation during radiological follow-up of reverse shoulder arthroplasty. The purpose of this study was to evaluate the effect of glenosphere design and humeral component retroversion on movement amplitude in the scapular plane and inferior scapular impingement. MATERIALS AND METHODS The Aequalis Reversed Shoulder Prosthesis (Tornier) was implanted into 40 cadaver shoulders. On the glenoid side, 8 different combinations were tested: 36-mm glenosphere: centered (standard), eccentric, with an inferior tilt, or with the center of rotation (COR) lateralized by 5 or 7 mm; and 42-mm centered glenosphere: used alone or with the COR lateralized by 7 or 10 mm. The humeral component was positioned in 0°, 10°, 20°, 30°, and 40° of retroversion. Maximum adduction and abduction were measured when inferior impingement and superior impingement, respectively, were detected. RESULTS The average increase in abduction amplitude was 10° and inferior impingement occurred 18° later with a 42-mm glenosphere, especially when it was lateralized by 10 mm, relative to a 36-mm centered glenosphere (P < .05). These 2 combinations provided a 28° increase in the movement amplitude in the scapular plane. Positioning of the humeral component in 10° or 20° of retroversion or in anatomical retroversion was most effective at avoiding inferior impingement but had less effect on abduction range of motion (except with the 42-mm glenosphere). CONCLUSION Our study confirmed published results with various glenosphere designs but was unique in describing the effect of humeral retroversion on scapular impingement. Inferior scapular notching can be most effectively prevented by using large-diameter glenospheres with lateralized COR and by making sure to replicate the patients native humeral retroversion.