Randomized phase II trial of erlotinib (E) versus temozolomide (TMZ) or BCNU in recurrent glioblastoma multiforme (GBM): EORTC 26034

2005

Background: In 40–50% of GBM epidermal growth factor receptor (EGFR) is amplified, and often constitutively activated (EGFRvIII mutant). EGFR is therefore a potential therapeutic target. Previous studies suggested activity of EGFR tyrosine-kinase inhibitors in recurrent GBM, particularly in specific molecular subsets. This study explored erlotinib (E) activity in recurrent GBM. Methods: Randomized phase II trial. Eligibility criteria: histologically proven GBM, recurrent >3 months after radiotherapy, Karnofsky performance status (KPS) =70, no prior chemotherapy for recurrent disease, tissue sample for EGFR studies. Patients (pts) received E 150mg/day (300mg/day if on enzyme inducing anti-epileptic drugs [EIAEDs]), or control (TMZ 150–200mg/m2, day 1–5 q4wk or BCNU 60–80mg/m2 i.v., day 1–3 q8wk). If no significant toxicity, E was escalated to 200mg (500mg in patients on EIAEDs). The primary endpoint was 6 months’ PFS in =10/50 pts on E (20%); P0 was set at 15%, and P1 at 30%. Response was assessed with Macdonald’s criteria. EGFR amplification (FISH) and expression of EGFR, EGFRvIII and PTEN (immunohistochemistry [IHC]) were assessed. Results: 110 patients were randomized (54 E, 56 control: 27 TMZ; 29 BCNU). Median age 55 years; median KPS 90. All but 1 patient started treatment; median number of cycles was 2 for E, 4 for TMZ and 1 for BCNU. Grade 3/4 toxicities likely related to E: dermatological (5); hemorrhage (1). Grade 3/4 toxicities for control were mainly hematological (3 TMZ, 13 BCNU). Three pts discontinued E due to toxicity. Six-month PFS was 12% for E, 24% for control. Six and 12-month survival were 61% and 24% for E, and 63% and 26% for control. Two responses were seen on control; the best response on E was SD (n=6). Patients with EGFRvIII mutations (13 in E arm, 8 in control arm) had shorter PFS (p=0.007) and OS (p=0.004) regardless of treatment. Response to E was not correlated with EGFR expression, EGFR amplification or EGFRvIII mutation. Conclusions: This randomized, controlled phase II study did not find sufficient activity for erlotinib in the general population of recurrent GBM. The presence of EGFRvIII mutations was not predictive for response.

No significant financial relationships to disclose.

EORTC Study 26041–22041: Phase I/II study on concomitant and adjuvant temozolomide (TMZ) and radiotherapy (RT) with or without PTK787/ZK222584 (PTK/ZK) in newly diagnosed glioblastoma—Results of a phase I trial

2026

Background: VEGF is an essential mediator of angiogenesis. PTK/ZK is an orally available tyrosine kinase inhibitor of all known VEGF receptors. The EORTC has initiated a randomized phase I/II study in which PTK/ZK is being investigated in combination with standard RT and concomitant TMZ. The results of the safety and dose-finding run-in of the triple combination are reported here. Methods: Escalating doses of PTK/ZK QD continuously (using classical 3+3 design) and concomitant RT (60 Gy) and TMZ (75 mg/m2/day) during 6–7 weeks were to be administered. During the adjuvant/maintenance part, PTK/ZK was prescribed at a fixed standard dose of 750mg bid until progression. Dose limiting toxicities (DLT) were defined as >G3 toxicity occurring during RT + 2 weeks. Results: 18 GBM pts (M/F: 11/7, median age: 53 years, PS 0/1) have been enrolled and complete data for 15 pts are currently available. DL1 (500 mg): 4 pts; no DLT; DL2 (1000 mg): 6 pts: 1 DLT: G3, liver enzyme (ALAT) elevation and hyponatremia DL3 (1250 mg); 8 pts: 3 DLTs: G3 hepatic toxicity (2 pts) and G3 platelets and G4 neutrophil (1 pt). In a 4th patient a G3 diarrhea and hepatic DLT leading to PTK/ZK interruption was doubtful therefore it was decided to extend DL3. Frequently observed possibly related non-dose limiting drug adverse events of all grades included: abdominal pain/cramping (2 pts), ALAT/ASAT elevation (10 pts), creatinine elevation (2 pts), allergic reaction (2 pts), constipation (4 pts), diarrhea (5 pts), fatigue (11 pts), hypertension (6 pts), nausea (8 pts), vomiting (3 pts), and weight loss (4 pts). All toxicities were reversible. Conclusions: MTD has been reached at 1250 mg and the recommended dose of PTK/ZK in combination with RT and TMZ is 1000 mg/day. A randomized 3-arm phase II will start accrual in January 2007. Patients will receive standard TMZ/RT alone or with PTK/ZK starting at the beginning of radiotherapy or at the beginning of adjuvant therapy 4 weeks after the end of radiotherapy.

[Table: see text]

Time finding study of chronomodulated irinotecan (I), fluorouracil (F), leucovorin (L) and oxaliplatin (O) (chronoIFLO) against metastatic colorectal cancer: Results from randomized EORTC 05011 trial

2566

Background: We previously showed that a) irinotecan (I) could be combined with chronoFLO in MMC patients; b) least toxic times (LTT) for combined I and oxaliplatin respectively correspond to the middle of the rest-phase and the middle of acivity-phase in tumor-bearing mice; c) chrono I showed adequate activity in a randomized phase II trial; and d) MMC resistance can be partly overcome with chronoIFL. Methods: The objective was to identify the LTT for I characterized by a minimal dose reduction/delay among the first 3 courses (c). Assuming that the toxic effect of I had a 24-h periodicity patients were randomized in 6 groups with I peak delivery (180 mg/m2, 6-h sinusoidal infusion on day 1) at 1:00, 5:00, 9:00 am, 1:00, 5:00, or 9:00 pm. . All the groups received also chronoFLO on days 2–5, q 3 weeks (F 700 mg/m2/d & L 150 mg/m2/d; from 22:15 to 9:45 with peak delivery at 4:00 , O 20 mg/m2/d from 10:15 to 21:45, with peak delivery at 16:00). Based on a logistic regression model, a 15% reduction in toxic events in the first 3 c, 186 patients were considered necessary to estimate the LTT with a 95% CI (calculated by bootsrap) of less than 6 h. Results: 197 of 199 randomized MMC patients were considered for tolerability and safety with median age 61 years (30–81), sex (M 68% - F 32%) and PS (0/1/2 73/23/4%); therapy was 1st line in 77 patients and 2nd line in 23%. Thithy-one percent of severe protocol violations occurred, 16% of pump malfunctions (>10% dose delivery deviation). Median number of c was 6 (1–18). There were 3 toxic deaths. The observed LTT for I tolerability was 3:15 am (95 CI: 3:40–1:50 pm, NS). Grade 3–4 diarrhea ranged from 34 to 51.6% with LTT at 1:53 pm (4:29 -2:53 am, not significant, NS); neutropenia from 9 to 25% with LTT at 3:26 pm (10:50 - 4:55 am, NS). Age was a negative prognostic factor for diarrhea (p =0.01). Conclusion: This trial failed to show a statistically significant LTT for this combination in MCC patients. The safety profile of I combined with ChronoIFLO was acceptable, with diarrhea and neutropenia within previously reported range.

No significant financial relationships to disclose.

Altered circadian rhythm in rest and activity (CircAct): Independent prognostic value for survival in patients (pts) with metastatic colorectal cancer (MCC)

14573

Background: Altered CircAct accelerated tumor growth in experimental models (JNCI 2002, 2005; Cancer Res 2004). CircAct was found to be related to survival in two studies (St) in MCC pts (St 1, Clin Cancer Res 2000; St 2, ASCO 2005 #3553). The current study attempted to provide an independent prognostic stratification criteria for survival according to baseline CircAct as a first step toward the development of a comprehensive prognostic model based on circadian physiology. Methods: CircAct was continuously and non- invasively assessed over 72h with a wrist accelerometer (Ambulatory Monitoring, USA), before chemotherapy in two independent datasets of 169 MCC pts (60% pretreated) at Paul Brousse Hospital (St 1) and 130 chemo-naïve MCC pts registered in the international EORTC 05963 trial (St 2). Activity distribution during rest and wakefulness was objectively quantified by I<O (50=no rhythm; 100=marked rhythm). Kaplan-Meier survival curves with logrank tests and Cox regression models were used for analyses. Results: In both St, pts with I<O = 92 (21% in St 1 and 22% in St 2) displayed a significantly poorer survival as compared to those pts with I<O > 92 (Table). The hazard ratios (HR) were 0.37 [95%CI: 0.25 to 0.55] (p<0.0001) in St 1 and 0.56 [0.36 to 0.87] (p=0.01) in St 2. The prognostic value of altered CircAct persisted after adjusting for PS (HR=0.46 [0.30 to 0.69], p=0.0002 in St 1; HR=0.61 [0.38 to 0.97], p=0.04 in St 2). Conclusions: The independent stratification of MCC pts for survival according to baseline altered CircAct occurred using a threshold value of 92 for I<O in both separate datasets. Subsequent studies will involve additional biomarkers of circadian physiology (cortisol, melatonin, clock genes) in order to further improve this prognostic stratification as a step toward tailored treatment of MCC pts.

[Table: see text]

No significant financial relationships to disclose.

Continuous versus intermittent tamoxifen versus intermittent/alternated tamoxifen and medroxyprogesterone acetate as first line endocrine treatment in advanced breast cancer: an EORTC phase III study (10863)

BACKGROUND

Continuous ligand depletion of endocrine responsive tumours may enhance resistance to therapy. Intermittent treatment with tamoxifen (T) was considered to mimic (incomplete) ligand depletion and reintroduction. Furthermore it was postulated that alternating tamoxifen with a non-cross resistant endocrine modality could (further) postpone hormone resistance.

PATIENTS AND METHODS

Postmenopausal patients with advanced breast cancer who did not progress after 4 months of first line T therapy were randomised to continue T (40 mg daily) or to 2 monthly intermittent T or intermittent/alternated T and medroxyprogesterone acetate (MPA, 300 mg daily). At progression during break or during MPA, T should be reintroduced. Endpoints of the study were progression free survival (PFS), time to resistance to tamoxifen and overall survival (OS).

RESULTS

Of 593 registered patients, 276 were randomised. After 8 years follow-up the median PFS for continuous T, intermittent T and intermittent/alternated T and MPA was 11.0 (8.1-15.2), 8.0 (6.2-12.4) and 10.8 (7.1-16.7) months, respectively (NS). Resistance to tamoxifen was established only in 84%, 70% and 55% of patients in the three treatment arms, respectively. The median times from randomisation to resistance to tamoxifen were 12.5 (9.1-21.1), 13.2 (8.8-19.8) and 24.0 (16.9-60.9) months, respectively (p<0.001), without translation in differences in survival times.

CONCLUSION

Intermittent T or intermittent/alternated T and MPA had no impact on PFS or OS as compared with classical continuous T in patients with advanced breast cancer. Intermittent/alternated T and MPA resulted in prolonged time to resistance to T, but this might partly be due to bias by omittance of the proof of tamoxifen resistance in a high proportion of the patients in this treatment arm.

Optimal circadian time of vinorelbine (V) combined with chronomodulated 5-FU in pretreated metastatic breast cancer patients. EORTC 05971 randomized multicenter study

2066

Background: Chronotherapy is an aim to increase efficacy/toxicity ratio. Objectives: to define the dosing least toxic time (DLTT) of V (30 mg/m2/d at D1 and D6), combined with chrono 5-FU (10 pm-10 am) (850 mg/m2 D2-D5) over 3 courses q3w. Methods: A logistic regression model (LRM) estimated the DLTT assuming a sinusoidal distribution over time (i.e. over the 8 different arms) of the toxicity rate observed in each arm. The associated 90% confidence limits (CL) has been obtained by bootstrap method. Results: 90 patients were recruited. Toxicity in 46 pts led to the V dosage reduction to 25 mg/m2/d. 40 and 43 pts were assigned the V30 and the V25 regimen. 12% pts went off for toxicity, 5% for PD, 1% for refusal, 1% for unrelated death. 224 cycles were analyzed . V and 5FU relative dose intensities were 79.4% and 78.2% in the V30 while 88.1% and 87.4% in the V25 pts. Over the 3 cycles, toxicity by cycle was: Grade (G) 3 and G4 leucopenia in 47% and 29%, G3 and G4 neutropenia in 12% and 77%. G3 febrile neutropenia in 34%. G2 thrombopenia and anemia in 4% each. Other G3 and G 4 toxicity were stomatitis (12%), alopecia (7%), and less than 5%: cardiovascular, lethargy, diarrhea, constipation, other gastrointestinal, infection, sensory, pulmonary. LRM could not demonstrate a DLTT for the neutropenia G3–4 incidence, the primary endpoint. However, based on the stratified by dose analysis, a 90% CI of less than 6 hours width was observed: - around 17H17 [14H04–20H03] for the incidence of leucopenia G3–4. - around 8H16 [06H04–10H39] for tolerability (dose reduction, dose delay or treatment interruption for toxicity reason). This suggests that treatment tolerability was influenced by other factors beside leucopenia nadir. No other 90% CI of less than 6 hours width could be observed for other toxicity endpoints. Conclusions: Using a novel time finding study design with ad hoc statistics, this first randomized multicenter study has determined a DLTT for Vinorelbine in 90 women with MBC. Additional studies are ongoing to further assess the relevance of this trial design method that could prove useful for improving the safety of anticancer drugs during their clinical development. Support Pierre Fabre Oncology, Ligue Bourguignonne contre le Cancer

No significant financial relationships to disclose.

A randomized phase III multicenter trial of neoadjuvant docetaxel plus cisplatin and 5-fluorouracil (TPF) versus neoadjuvant PF in patients with locally advanced unresectable squamous cell carcinoma of the head and neck (SCCHN). Final analysis of EORTC 24971

5516

Background: So far, the Wayne State regimen combining 100 mg/m2 cisplatin on day 1 and 1000 mg/m2/d continuous infusion 5-fluorouracil over 5 days (PF), is the gold standard in advanced SCCHN. Improvement of PF induction chemotherapy by adding a taxane has been suggested from phase II studies and 2 randomized phase III trials. We now report the final analysis of EORTC 24971. Methods: Eligible were patients (pts) with unresectable stage III/IV SCCHN (no distant metastases), 18 to 70 yrs of age, PS≤1, and adequate hematologic, renal and hepatic function. Pts were stratified by primary disease site (oral cavity, oropharynx, hypopharynx, larynx) and treatment center and randomized to PF (as above) or TPF (T 75 mg/m2/1h, P 75 mg/m2/1h, both d1, F 750 mg/m2/d, d1–5) q 3 wks, for 4 cycles unless progression (PD) or unacceptable toxicity, or refusal. Thereafter (interval 4–7 wks), all pts not in PD received radiotherapy (RT: conventional [66–70 Gy], accelerated [max. 70 Gy] / hyperfractionated [max. 74 Gy]). Surgery was allowed before RT (neck) or 3 months after RT (primary, neck). Primary endpoint was progression-free survival (PFS), and 348 pts and 260 events were needed to detect a 50% increase in median PFS (10 vs 15 mo) with 85% power. Results: Between April 1999 and March 2002, 358 pts were accrued (181 PF, 177 TPF). At a median follow-up (FUP) of 32 mo, PFS was significantly improved with TPF (HR 0.72, p = 0.0071; median 8.2 vs 11.0 mo). A 51 mo median FUP showed an overall survival (OS) benefit with TPF (HR 0.71, p = 0.0052; median 14.2 vs 18.6 mo). Estimated 3-yr survival rates are 23.9% (95% CI: 17.9%;30.5%) for PF and 36.5% (29.3%;43.6%) for TPF. Response rate also favored TPF (53.6% vs 67.8%, p = 0.006). There was more severe (G3+4) leucopenia (41.6% vs 22.9%) and neutropenia (76.9% vs 52.5%) with TPF, and more severe thrombocytopenia with PF (17.9% vs 5.2%). Severe alopecia occured more with TPF (55.5% vs 11.7%); hearing loss (2.8% vs 0%) and toxic death (7.8% vs 3.7%) more with PF. Conclusions: TPF (→ RT) is superior to PF (→ RT) for PFS, OS (including long-term), response rate, and is better tolerated.

[Table: see text]

Phase II study of first-line chemotherapy with temozolomide in recurrent oligodendroglial tumors: the European Organization for Research and Treatment of Cancer Brain Tumor Group Study 26971

PURPOSE

Oligodendroglial tumors are chemotherapy-sensitive tumors, with two thirds of patients responding to combination chemotherapy with procarbazine, lomustine, and vincristine (PCV). Temozolomide (TMZ), a new alkylating and methylating agent, has demonstrated high response rates in patients with recurrent anaplastic astrocytoma. We investigated TMZ as first-line chemotherapy in recurrent oligodendroglial tumors (OD) and mixed oligoastrocytomas (OA) after surgery and radiation therapy.

PATIENTS AND METHODS

In a prospective, nonrandomized, multicenter, phase II trial, patients were treated with 200 mg/m2 of TMZ on days 1 through 5 in 28-day cycles for 12 cycles. Patients with a recurrence after prior surgery and radiotherapy, and with measurable and enhancing disease on magnetic resonance imaging (MRI) were eligible for this study. Patients with large lesions and mass effect or with new clinical deficits were not eligible. Pathology and the MRI scans of all responding patients were centrally reviewed.

RESULTS

Thirty-eight eligible patients were included. In three patients, pathology review did not confirm the presence of an OD or OA. TMZ was generally well tolerated. The most frequent side effects were hematologic; only one patient discontinued treatment for toxicity. In 20 (52.6%) of 38 patients (95% exact confidence interval, 35.8% to 69.0%), a complete (n = 10) or partial response to TMZ was observed. The median time to progression was 10.4 months for all patients and 13.2 months for responding patients. At 12 months from the start of treatment, 40% of patients were still free from progression.

CONCLUSION

TMZ provides an excellent response rate with good tolerability in chemotherapy-naive patients with recurrent OD. A randomized phase III study comparing PCV with TMZ is warranted.

Economic evaluation of antibiotic prophylaxis in small-cell lung cancer patients receiving chemotherapy: an EORTC double-blind placebo-controlled phase III study (08923)

BACKGROUND

To determine whether the cost of prophylactic antibiotics during chemotherapy is offset by cost savings due to a decreased incidence of febrile leukopenia (FL).

PATIENTS AND METHODS

Small-cell lung cancer (SCLC) patients were randomised to standard or intensified chemotherapy with granulocyte colony-stimulating factor to assess the impact on survival (n = 244). In addition, patients were randomised to prophylactic ciprofloxacin and roxithromycin or placebo to assess the impact on FL (n = 161). The economic evaluation examined the costs and effects of patients taking antibiotics versus placebo. Medical resource utilisation was documented prospectively, including 33 patients from one centre in The Netherlands (NL) and 49 patients from one centre in Germany (GE). The evaluation takes the perspective of the health insurance systems and of the hospitals. Sensitivity analyses were performed.

RESULTS

In the main trial, prophylactic antibiotics reduced the incidence of FL, hospitalisation due to FL and use of therapeutic antibiotics by 50%. In GE, the incidence of FL was not reduced by prophylaxis. This resulted in an average cost difference of only 35 Euros [95% confidence interval (CI) (-)1.713-2.263] in favour of prophylaxis (not significant). In NL, prophylaxis reduced the incidence of FL by nearly 50%, comparable with the results of the main trial, resulting in a cost difference of 2706 Euros [95% CI 810-5948], demonstrating savings in favour of prophylactic antibiotics of nearly 45%. Sensitivity analyses indicate that with an efficacy of prophylaxis of 50%, and with expected costs of antibiotic prophylaxis of 500 Euros or less, cost savings will incur over a broad range of baseline risks for FL; that is, a risk >10-20% for FL per cycle.

CONCLUSIONS

Giving oral prophylactic antibiotics to SCLC patients undergoing chemotherapy is the dominant strategy in both GE and NL, demonstrating both cost-savings and superior efficacy. The sensitivity analyses demonstrate that, due to the efficacy of prophylactic antibiotics and their low unit cost, cost savings will incur over a broad range of baseline risks for FL. We recommend the use of prophylactic antibiotics in patients at risk for FL during chemotherapy.

Vinorelbine and cisplatin in metastatic squamous cell carcinoma of the oesophagus: response, toxicity, quality of life and survival

BACKGROUND

Vinorelbine and cisplatin are active against squamous cell oesophageal carcinoma. The purpose of this phase II study was to evaluate the efficacy and safety of vinorelbine plus cisplatin in previously untreated patients with metastatic squamous cell oesophageal carcinoma and to estimate the progression-free survival, overall survival and quality of life (QoL) of the patient population.

PATIENTS AND METHODS

Seventy-one eligible patients were entered into a study of vinorelbine 25 mg/m2 on days 1 and 8 plus cisplatin 80 mg/m2 on day 1, every 3 weeks. Degree of dysphagia relief was monitored and QoL was measured using the EORTC QLQ-C30.

RESULTS

All eligible patients were assessed for response and 24 achieved a confirmed partial response (33.8%; 95% confidence interval 23-46); the median duration of response was 6.8 months, progression-free survival was 3.6 months and median survival of the whole group was 6.8 months. Toxicity was mainly related to neutropenia (grade 3/4 in 41% of patients). At cycle 2, 43% of the patients reported at least a moderate improvement in global health status/QoL and 25% experienced a large improvement.

CONCLUSIONS

Vinorelbine plus cisplatin represents a well-tolerated active palliative regimen for patients with advanced squamous cell carcinoma of the oesophagus. This combination may offer a better therapeutic index than cisplatin-5-fluorouracil.

Estimating survival gain for economic evaluations with survival time as principal endpoint: a cost-effectiveness analysis of adding early hormonal therapy to radiotherapy in patients with locally advanced prostate cancer

The problem of estimating expected outcomes for the economic evaluation of treatments for which the outcome of principal interest is (quality adjusted) survival time has so far not received sufficient attention in the literature. The best estimate of expected survival is mean survival time, but with censored survival data, the true survival time for all the subjects is not known, so the mean is not defined.A possible solution to this estimation problem is illustrated by a retrospective cost-effectiveness analysis of the addition of hormonal therapy to standard radiotherapy for patients with locally advanced prostate cancer. A recently proposed method is used to approach the problem caused by censored cost data, and the impact of uncertainty is assessed by bootstrap resampling techniques. Mean survival time is estimated by a restricted means analysis with the time point of restriction determined by statistical criteria. When average total costs and mean survival time is evaluated at this time point of restriction, the result is that the combined therapy (radiotherapy plus hormonal therapy) increases mean survival time by about 1 year, while reducing the costs per patient for the French health insurance system by 12 700 FF. The time point of restriction may also be determined by other criteria and mean survival time may be estimated by extrapolating the survival curves by means of various parametric survival distributions. We show that the exact results of the economic evaluation are decisively determined by the restriction time point chosen and the approach taken to estimate mean survival time.

Statistical methodology of phase III cancer clinical trials: advances and future perspectives

The methodology for conducting cancer clinical trials has undergone enormous changes over the past 25-30 years since the EORTC Data Center was created. The purpose of this paper is to highlight and to provide a historical perspective for the main methodological concepts, both practical and theoretical, which form the basis for the design and analysis of phase III cancer clinical trials within the EORTC Data Center. Some statistical aspects of other associated topics such as quality of life, health economics, meta-analysis and treatment outcome will also be briefly discussed. Finally, some future perspectives and topics for further statistical methodological research will be presented in order to spur statisticians to meet the challenge of efficiently designing and analysing the clinical trials of tomorrow.

Cost-effectiveness of the addition of early hormonal therapy in locally advanced prostate cancer: results decisively determined by the cut-off time-point chosen for the analysis

We present a retrospective cost-effectiveness analysis using data from a randomised controlled trial (EORTC 22863) of the addition of early hormonal therapy with a luteinising hormone-releasing hormone (LHRH) analogue to radiotherapy in the treatment of patients with locally advanced prostate cancer. Data on the use of medical resources were extracted from the hospital charts of 90 patients recruited into the trial by one French hospital. Costs are assessed from the viewpoint of the French healthcare financing system and adjusted for censoring. Expected costs per patient of each treatment is related to the expected outcome, mean survival time, estimated by a restricted means analysis. The time point of restriction is determined by statistical criteria. In the base case analysis with a cut-off time point at 8.58 years, the combined therapy group (COMB) had a gain in mean survival time of 1.06 years (7.05 versus 5.99 years) and a reduction of average total costs of 12700 French francs (FF) (58300 FF versus 71000 FF). The analysis of uncertainty uses bootstrap techniques with 5000 replicates to examine the joint distribution of cost and survival outcomes. In 76% of the cases, COMB results in longer mean survival time and lower costs than the radiotherapy group (RT). In cases where COMB therapy raises costs (13% of the cases), it is rarely by more than 20000 FF per patient, no matter the size of the associated survival gain. It is thus highly likely that COMB should be considered a cost-effective option compared with RT for these patients. The exact result of the economic evaluation is decisively determined by the restriction time point selected for the determination of mean survival time, partly also because the average total costs of the two treatments develop entirely differently as a function of the survival time.