Prognostic role of pregnancy occurring before or after treatment of early breast cancer patients aged <35 years: a GET(N)A Working Group analysis

BACKGROUND

Usual practices recommend waiting at least 2 years between diagnosis of early breast cancer (EBC) and pregnancy. Few data highlighted a harmful effect of an early pregnancy for low-risk patients. The authors analyzed retrospectively data from women younger than 35 years who became pregnant before or after treatment of EBC.

METHODS

Between 1990 and 1999, 908 consecutive EBC patients were analyzed. The primary endpoint was to compare overall survival (OS) between pregnant and nonpregnant patients. The secondary endpoint was to establish a score index laying down the risk of distant recurrence.

RESULTS

Within the year before the diagnosis, 105 (11.6%) patients became pregnant and 118 (13%) were pregnant after treatment. In a multivariate model, a pregnancy before the diagnosis was not predictive of death but of local relapse. A pregnancy subsequent to breast cancer therapy resulted in a 77% decrease of death (P < .001). In good-prognosis score index patients, the annual risk of relapse remained low. In patients having the higher score, recurrences occurred mainly during the first years after the treatment. Beyond 80 months, the annual risk of relapse seemed to be similar to those of lower-risk subgroups.

CONCLUSIONS

In women aged younger than 35 years, a pregnancy occurring before or after the diagnosis of breast cancer was not an independent prognostic factor of death. In the subset of patients having a high risk of relapse, it may be preferable to postpone a pregnancy beyond 5 years after the breast cancer therapy.

Docetaxel and pegylated liposomal doxorubicin combination as first-line therapy for metastatic breast cancer patients: results of the phase II GINECO trial CAPYTTOLE

BACKGROUND

This phase II study evaluated the clinical benefit of pegylated liposomal doxorubicin (PLD) and docetaxel (Taxotere) as first-line therapy for metastatic breast cancer (MBC).

PATIENTS AND METHODS

MBC patients were enrolled to receive six cycles of PLD 35 mg/m2 (day 1) and docetaxel 40 mg/m2 (days 1 and 15), every 28 days (group A). Because of unacceptable toxic effects, doses were modified to PLD 30 mg/m2 (day 1) and docetaxel 75 mg/m2 (day 2), every 3 weeks (group B). The primary end point was clinical benefit.

RESULTS

Sixty-seven patients were included (group A, 53; group B, 14). In both groups, the median number of cycles delivered was 4 and the overall dose intensity was 82% for docetaxel and 71% for PLD. In group A, main toxic effects were hematologic, palmar-plantar erythrodysesthesia (PPE), and stomatitis. In group B, higher rates of grade 3-4 PPE, febrile neutropenia, and hematologic toxic effects were reported. The rate of clinical benefit was 47%. Among patients with a measurable disease, 49% achieved a partial response, 27% had a stable disease, and 13% progressed, according to RECIST criteria.

CONCLUSION

The combination of PLD and docetaxel delivered at planned doses in this study yields unacceptable toxicity and should not be used routinely in patients with MBC.

Women's perceptions and experience of adjuvant tamoxifen therapy account for their adherence: breast cancer patients' point of view

OBJECTIVE

The aim of this study on primary breast cancer patients undergoing adjuvant tamoxifen treatment was to determine how their perceptions of the treatment and their experience of side-effects contributed to their adherence to the treatment.

METHODS

A consecutive series of primary breast cancer patients eligible for tamoxifen therapy were studied qualitatively by conducting semi-structured in-depth interviews at two French cancer centres.

RESULTS

The women aged 35-65 (N=34) were struggling with several issues involving their understanding and experience of the treatment, which have not been documented so far. These issues included confusion about the 'hormonal' nature and activity of tamoxifen and the etiology of the changes in their menopausal status, as well as the symbolic associations formed by patients about the paradox of taking a treatment that has aging effects but saves lives.

CONCLUSIONS

This study shows the great physical burden often associated with tamoxifen treatment and brings to light women's own complex representations of the treatment and their interpretation of the side-effects. Better communication between health-care providers and patients should ultimately help to prevent refusal or discontinuation of tamoxifen treatment.

Distal or proximal venous port device insertion: Results of a prospective randomized trial

e20605

Background: Open, nonblinded, prospective, randomized, controlled trial comparing two techniques of venous port device insertion: percutaneous distal (phlebography-guided arm port- study technique- 2) vs proximal surgical (cephalic vein cutdown- control technique- 1) placement -To determine whether technique 1 is superior to technique 2. Methods: 230 eligible patients beginning a course of i.v. chemotherapy via a port device catheter with an expected duration of treatment of 3 months or longer were randomized (written informed consent). Eligibility criteria included adult patients with solid tissue malignancy (neuro oncology, gynecology, lung, abdominal, head§neck) beginning a course of I.V.chemotherapy, normal hemostatic parameters, no organ failure, a life expectancy >3months, WHO status<3. Exclusion criteria included current anticoagulant therapy, previous ipsilateral venous catheter/pacewires/surgical axillary node dissection/radiodermatitis, local tumor growth/sepsis, symptomatic brain metastasis, psychosis. The silicone rubber 7F catheter was connected to a 11mm port reservoir, and implanted under local anesthesia using either technique 1 or 2 after randomization. Outcome measurements included technical feasibility/procedure duration, port complications, quality of life (EORTC) questionnaires. Results: Median study duration was 12.2 vs 11.9 months (p: 0.9), median chemotherapy cycles were 6.0 in both groups. Patients groups differed significantly in venous access side (left access in group 2) and sex ratio (p=0.028). In group 2, technical success rate was higher (99 vs 91%, p<0.02), procedure was shorter: 18.0min (10.0–90.0) vs 21min (15.0- 45.0)(p<0.008), but global complication rate was higher (p<0.05). Device complication related explantation rate was 11.9 vs 2.8% (p=0.022). Conclusions: Both techniques are safe and effective. Despite a higher technical success rate and a shorter procedure duration, arm port insertion has a lower complication-free duration. Distal (arm port) technique should be recommended in young female cancer patients (neckline cosmesis/discretion), head and neck cancer patients, obese patients (upright position) and in patients presenting with respiratory insufficiency or at high risk for pneumothorax.

No significant financial relationships to disclose.

Weekly paclitaxel and capecitabine in HER2-negative metastatic breast cancer (MBC): A multicenter GINECO randomized phase II comparing two paclitaxel-capecitabine schedules

1114

Background: Paclitaxel (Ta) and capecitabine (Xel) are synergistic in vitro. Compared to a continuous weekly Tax combined with a classical 14 days (d)/21 Xel administration (Elza-Brown et al., ASCO 2000), we have explored the combination of Xel 5d/week (wk) and weekly Ta 3 wk out 4 in the objective to increase the efficacy/toxicity ratio of the TaXel combination. Methods: Patients (pts) in first or second-line of MBC, previously treated with anthracyclins + docetaxel were randomized either to A: Ta (60 mg/m2/w) + Xel (2000 mg/m2/d x 14 d/21) or to B: Ta (80 mg/m2/w) + Xel (2000 mg/m2/d x 5 d/wk) 3wk out 4. Results: From January 2006 to January 2008, 130 pts were accrued (A 66, B 64). Pts characteristics were well balanced between the two arms including median age (58 yrs), histologic type and grade, hormone receptor-positive tumor (80%), previous treatment, visceral disease (72%), number of sites (>1; 63%), ECOG PS (0; 42%, 1; 58%). Pts received a median of 6 cycles (1–23) with a received/planned mean dose of 89.3% for Ta in both arms and of 74 and 76% for Xel respectively in arm A and B. Haematological toxicity (Tox) was low in both arms with neutropenia Gr 3 in only 8% of cycles, G-CSF support in 2% of cycles, and infection G3 in 5 pts. Alopecia G2 was less frequent in arm A (29 vs 60%). Other Tox were similar in both arms: [G2/3 (%) cutaneous (35/17), pain (36/9), fatigue (26/13), neuropathy (20/3), diarrhea (15/6), mucositis (8/2), vomiting (9/1)] but treatment interruption due to Tox was more frequent in A (A 19, B 7 pts) (p = 0.02). Response rate was 52% (B) versus 44% (A). A progression-free survival advantage was seen for B over A (366 vs 272 days, p = 0,15) including in the triple negative pt subset (n = 26 pts) (197 vs 150 days, p = 0.07). Conclusions: The intermittent regimen (3 wk out 4) of weekly paclitaxel and capecitabine 5 d/week is a well accepted, safe and effective TaXel schedule and might be a chemotherapy regimen of choice in MBC including triple negative patients.

No significant financial relationships to disclose.

Weekly paclitaxel (wP) as single agent or in combination with weekly topotecan (wT) or carboplatin (C) in patients with resistant ovarian cancer (ROC): The phase II CARTAXHY randomized trial from GINECO

5557

Background: For ROC patients (pts) with early progression during or after (< 6 months) platinum and 3 weekly P, use of single non-platinum agent including wP is standard (Kristensen G, et al. J Clin Oncol 26: 2008 abstr 5508). Few randomized trials have explored combination therapy in this setting. Methods: Pts with ROC after a first or second line including a platinum and a taxane were randomized to receive wP (80 mg/m2/week) alone or in combination with wT (3mg/m2/week) or C (AUC 5 every 4 weeks). The primary end-point was the comparison of progression-free survival (PFS) between single non-platinum agent and combination therapy (wP+wT or wP+C). Secondary objectives included safety, QoL, response rate (RR) and overall survival. Results: From April 2004 to August 2008, 165 pts were accrued (wP 57, wP+wT 57, wP+C 51). Median number of cycles and P dose-intensity (mg/m2/week) was 4.6 and 70, 4.2 and 63 in monotherapy and combination therapy arms respectively. Non-hematological toxicity was not different between the arms, except an excess of hypersensitivity reactions in the wP+C arm. Grade 3–4 neutropenia (48 vs 13% of pts), and anemia (24 vs 6%) were more frequent in combination therapy than in single agent arm and similar with wT or C combination. Febrile neutropenia was experienced by 5 pts treated with combination therapy. Discontinuation from drug treatment was more frequent with combination therapy (24% of pts) than with monotherapy (4%), mainly due to hematotoxicity. RR was 34, 38 and 39% for wP, wP+wT and wP+C respectively. Median PFS of pts treated with single agent (112 days) was not significantly different from those treated with combination therapy (149 days) (p = 0,62) and was similar in wP+wT (152 days) or wP+C (146 days) arms. Treatment with single non-platinum agent or combination therapy was not found an independent parameter when added to a Cox model including prognostic variables. Conclusions: Combination therapy (CT) in platinum resistant ovarian cancer was found more toxic than weekly paclitaxel and the PFS advantage from CT was not statistically significant.

No significant financial relationships to disclose.

Aromatase inhibition in male breast cancer patients: Biological and clinical implications

1130

Background: Because male breast cancer (MBC) is rare, treatment recommendations are derived from results of trials in female patients (pts). Although, several studies have shown the superiority of third-generation aromatase inhibitors (AI) over tamoxifen in menopausal women with advanced breast cancer, the role of such molecules remains unknown in male pts. We report here the largest experience about the efficacy of AI in MBC pts with advanced disease and their impact on estradiol (E) levels. Methods: MBC pts were selected from the breast cancer database of the Centre Antoine-Lacassagne (Nice, France) as follows: Metastatic disease with at least one measurable or assessable non-measurable lesion, estrogen receptor (ER) and/or or progesterone receptor (PR) positive, availability of complete clinical and histological data, evidence of progressive disease at initiation of AI, receipt of at least one month of treatment with non steroidal (anastrozole, letrozole) or steroidal (exemestane) AI. Sex hormone levels were retrospectively assessed on serum samples from our institutional serum bank. Results: 15 pts entered the study. Median age was 68 (range 39–85). 7 pts received previous lines of hormonal therapy (median = 1) and 3 pts a previous line of chemotherapy before the introduction of AI. The best response was complete response in 2 pts (13%), partial response in 4 pts (27%), stable disease (SD) in 2 pts (13%) and progressive disease (PD) in 7 pts (47%). The median duration of objective response (OR) was 11.6 months (95% confidence interval [CI] 7.6–15.5). At the time of analysis, 8 pts (53%) had died and 7 (47%) were still alive. The median PFS and OS were 4.4 months (95% CI 0.1–8.6) and 33 months (95% CI 18.4–47.6) respectively. The 1-year PFS and OS rates were 20% (95% CI 9.7–30.3) and 84.6% (95% CI 74.6–94.6), respectively. 9 out of 11 pts with available samples had E levels less than the lower limit of the assay during AI treatment. Among the 9 pts with E level decrease, four had OR, one had SD, and four had PD. 1 pt had E levels higher than the upper limit of the assay during AI treatment. This pt showed disease progression. Further data on FSH and testosterone levels will be presented at the meeting. Conclusions: AI are active in MBC pts. This activity is correlated with a significant reduction in E levels.

No significant financial relationships to disclose.

Long-term follow-up and factors of survival of HER-2 positive breast cancer patients treated either by neoadjuvant trastuzumab docetaxel (TAXHER-S01 study) or by neoadjuvant trastuzumab docetaxel carboplatin (GETN[A]1 study)

e11507

Background: Almost 20% of breast cancers over express Her2, which is associated with a more aggressive phenotype and with a decreased survival. Nevertheless, trastuzumab (T) has been a revolutionary step in the adjuvant and in the metastatic treatments of Her2 positive breast cancers. Here, we focus on neoadjuvant T and try to determine the factors correlating with disease free survival and with overall survival in Her2 positive breast cancer treated with T based neoadjuvant chemotherapy. Methods: Data from two published T based neoadjuvant phases II were used: the TAX-HER trial which studied the use of 6 courses of 3 weekly docetaxel with weekly neoadjuvant T (scheme TH) (Coudert et. al. Annals of Oncology 2006) and the GET(N)A-1 trial which studied the use of 6 courses of 3 weekly docetaxel and carboplatin along with weekly neoadjuvant T (scheme TCH) followed by 3 weekly adjuvant T (Coudert et. al. JCO 2007). Moreover, additional patients from our institution and treated by neoadjuvant TH and adjuvant T were included. Survival curves were estimated using Kaplan-Meier methods and compared by log-rank test. Results: Data was available for 128 patients. 62 patients (48.4%) received neoadjuvant TH from whom 39 did not receive adjuvant T. 66 (51.6%) received neoadjuvant TCH and adjuvant T. Tumors characteristics were as followed: 65 (50.7%) SBR 1–2, 54 (42.19%) SBR 3, 49 (38.28%) hormonal receptors (RH) negative and 72 (56.25%) RH positive. The rate of pathological complete response (pCR) (Chevalier 1/2) was 39.6%. Overall survival (OS) for the entire cohort was 74,8 months. Relapse was defined as local, regional, metastatic relapse or death. Survival without relapse (SR) was 74.8 months. No difference was noted in OS and in SR according to the type of chemotherapy, TH or TCH. pCR did significantly influence SR (p = 0. 03) and survival without local recurrence (SLR) (p = 0.04) but neither OS nor survival without metastatic relapse (SMR). Multivariate analysis demonstrated that OS was correlated with node response (as defined by sataloff grade NA or NB) (p=0.0275) and the use of hormonal therapy in RH positive tumors (p=0.0724).

[Table: see text]

Prognostic factors in 1,038 women with metastatic breast cancer

BACKGROUND

Treatment of metastatic breast cancer (MBC) remains palliative. Patients with MBC represent a heterogeneous group whose prognosis and outcome may be dependent on host factors. The purpose of the present study was dual: first, to draw up a list of factors easily available in everyday clinical practice requiring no sophisticated or costly methods and second, to provide results from a large cohort of women who underwent diagnostic and treatment at a single institution.

PATIENTS AND METHODS

From 1975 to 2005, a total of 1,038 women with MBC during their follow-up were included in this retrospective analysis. Patients were subsequently assigned to five groups according to the period of metastatic diagnosis.

RESULTS

It is shown that age at initial diagnosis, hormonal receptor status and site of metastasis are the most relevant prognostic factors for predicting survival from the time of metastastic occurrence. It is also shown that a metastasis-free interval is an easily and immediately available multifactorial prognostic index reflecting the multiparametric variability of the disease.

CONCLUSION

These fundamental observations may assist physicians in evaluating the survival potential of patients and in directing them toward the appropriate therapeutic decision.

Role of the HER2 [Ile655Val] genetic polymorphism in tumorogenesis and in the risk of trastuzumab-related cardiotoxicity

BACKGROUND

To examine the impact of a frequent her2 gene polymorphism (Ile655Val) on tumor growth and on the pharmacodynamics of treatment by trastuzumab.

PATIENTS AND METHODS

Experimental study: The growth characteristics of cells expressing the Ile or Val isoform were examined in vitro and after injection into nude mice. The effect of trastuzumab was determined in both experimental models. Clinical study: 61 patients with advanced breast cancers and treated by trastuzumab were genotyped for HER2 by PCR-RFLP. The influence of HER2 genotype on the trastuzumab treatment was examined.

RESULTS

Experimental study: HER2-expressing cells acquired the characteristics of tumor cells. The Val isoform-expressing cells showed the highest growth capacity and developed aggressive tumors sensitive to trastuzumab. Clinical study: There was no link between tumor response or survival and HER2 genotype. All cases of treatment-related cardiotoxicity were found in the Ile/Val group and there was no cardiac toxicity in the Val/Val and Ile/Ile patients.

CONCLUSIONS

This study establishes a clear-cut difference between the two HER2 isoforms regarding their tumorogenic potential with an advantage for the Val/HER2 isoform. In breast cancer patients treated with trastuzumab, the presence of a Val allele may constitute a risk factor for cardiac toxicity.

Decision-making and breast cancer clinical trialsHow experience challenges attitudes

OBJECTIVE

The aim of this study was to measure women's preferences about decision-making and their impact to participate or not to a hypothetical randomised controlled trial (RCT).

METHODS

We surveyed prospectively breast cancer patients invited to participate in a clinical RCT (group 1a=201 acceptances, group 1b=66 refusals) or not invited (group 2=188). All women had the same treatment.

RESULTS

Decision-making preferences of patients who had refused clinical RCT entry were more patient's centred (72.3%) compared to those of patients who accepted (35.0%, P<0.001). Altruism was not a significant determinant of patients' participation. Randomisation was considered acceptable in 52.0% (group 1a) compared to 16.9% and 21.1% for group 1b or group 2, respectively (P<0.001). It was the main predictor of willingness to participate in a hypothetical RCT (adjusted odds ratio (OR(adj)) 4.6; 95% confidence interval [2.7-7.7]; P<0.001) with the patient group allocation (OR(adj) group 1a=5.0 [2.9-8.7]; group 1b=0.2 [0.0-0.8]; group 2=1 [referent]; P<0.001). After multivariate adjustment, willingness to participate was also significantly related with medical decision-making preferences (OR(adj) 2.2 [1.0-4.9]; P=0.045), with the feeling of being unable to refuse a doctor's proposal (OR(adj) 1.8 [1.1-3.2]; P=0.031), and with satisfaction with doctors' communication (OR(adj) 3.1 [1.5-7.8]; P<0.001).

CONCLUSIONS

Patients' acceptance to participate in a RCT is preferred to be doctor's decision, whereas refusal is a personal one. When proposing a RCT, doctors must deal with patients' a priori negative feelings about randomisation. They should thoroughly discuss the reasons for and importance of randomisation as well as the other aspects of participating in the trial in order to give patients all of the information they need to make an informed decision.

Early versus late local recurrences after conservative treatment of breast carcinoma: differences in primary tumor characteristics and patient outcome

OBJECTIVES

To investigate whether some aspects of patient or tumor characteristics influence the timing of local recurrence (LR) in breast cancer treated conservatively, and to assess the impact of the timing of LR on patient outcome.

METHODS

A retrospective analysis was conducted on patients treated with conservative breast surgery followed by radiotherapy for breast carcinoma who developed LR. Out of 2,008 cases treated in our Institute between 1977 and 2002, 180 ipsilateral LR were observed. Of these, 46 LR were observed within 36 months after treatment, called early local recurrence (ELR), 44 developed between 37 and 60 months, called medium local recurrence (MLR), and 90 occurred after 60 months, called late local recurrence (LLR). Patient and tumor characteristics were analyzed in the 2 groups and compared.

RESULTS

Primary tumors >20 mm were more frequently found in patients with ELR (31%) than in patients with LLR (17%, p = 0.047). Grade 3 tumors were more often encountered in patients with ELR than in patients with LLR (27 versus 7%, p = 0.0002). Patients with ELR more frequently had tumors with negative estrogen receptors than patients with LLR (37% versus 6%, p < 0.0001). There was no statistically significant difference in the axillary lymph node (LN) status between patients with ELR and those with LLR (35 and 23% of positive LN, respectively, p = 0.24). Tumor size, grade, LN status, hormone receptors and the timing of LR affected the specific survival (SS) from initial surgery. On multivariate analysis, only LN status and the timing of LR retained an independent prognostic value, with an odds ratio of 6.7 for ELR. After LR, the SS was also influenced by all of the above factors, and on multivariate analysis, LN status, hormone receptors and the timing of LR were independent predictors with an odds ratio of SS of 2.50 in case of ELR (p = 0.006). The 5-year SS after LR for ELR, MLR and LLR were 55.8, 74.8 and 79.5%, respectively.

CONCLUSIONS

Unfavorable tumor characteristics such as big size, high grade, lack of hormone receptors, but not LN status, were associated with ELR. These findings suggest that patients with such aggressive tumor characteristics who do not recur early will have a lower risk of LLR than patients with more favorable factors.

Assessment of Care by Breast Cancer Patients Participating or Not Participating in a Randomized Controlled Trial: A Report With the Patients' Committee for Clinical Trials of the Ligue Nationale Contre le Cancer

Purpose

Cancer patients participating in randomized controlled trials (RCTs) have not been found to have better clinical outcomes than other patients. Our objective was to assess the impact of RCTs on patients' satisfaction with care.

Patients and Methods

A prospective study was carried out in a cohort of women with breast cancer (N = 455) divided into those invited to participate in an RCT (201 acceptances, 66 refusals) and a comparable control group not invited to participate (n = 188). All the patients underwent the same treatment (fluorouracil, epirubicin, and cyclophosphamide 100 mg/m2 for six cycles). One and 7 months after the beginning of chemotherapy, self-administered satisfaction scores were used to compare the women's assessment of their care (Comprehensive Assessment of Satisfaction with Care validated scale).

Results

At the beginning of chemotherapy, women to whom RCT had been proposed rated the doctors' availability (average ± standard deviation [SD]: RCT acceptance group, 3.60 ± 0.78; RCT refusal group, 3.68 ± 0.87; control group, 3.41 ± 0.82; P ≤ .02) and the doctors' communication (average ± SD: RCT acceptance group, 3.56 ± 0.88; RCT refusal group, 3.67 ± 0.88; control group, 3.39 ± 0.84; P ≤ .05) higher than those to whom the trial was not proposed. After the treatment, participants in the RCT felt that their doctor was more supportive (average ± SD: RCT acceptance group, 3.04 ± 0.92; control group, 2.77 ± 0.85; P = .005) and more informative about their illness and treatment (average ± SD: RCT acceptance group, 3.34 ± 0.88; control group, 3.08 ± 0.92; P = .006) than those in the control group. The general level of satisfaction was also higher in the RCT acceptance group.

Conclusion

Women participating in an RCT have a more positive picture of their doctors' care than others, probably because of the structural effects of the informed consent and data collection processes.

Concomitant weekly carboplatin (CB) and paclitaxel (P) with pelvic radiotherapy (RT) for the treatment of advanced cervical cancer (ACC): A FNCLCC gynecologic group phase I trial

5542

Background: Standard primary treatment for locally ACC is RT with concomitant chemotherapy (CT). CB and P are radiosensitizers with in vitro synergistic action. Methods: Patients (pts) with FIGO stage IIB-IVA negative paraaortic lymph nodes cervical cancer were treated with 6 weekly cycles of CT during pelvic RT (45 Gy) and brachytherapy (BT) according to Table 1 . Each dose escalation step followed a 30-day period of observation on cohorts of 3 to 6 pts depending on dose limiting toxicity (DLT): toxic death; garde (G) 4 neutropenia > 1 week; G 4 toxicity (other hematologic or non-hematologic); any toxicity requiring = 1 week delay in RT, or > 2 dose reductions of CT, or G 3/4 hematologic toxicity > 3 weeks after treatment’s end; unendurable G 3 non hematologic toxicity. Results: 23 pts were included by 5 centers in 5 dose levels (L). Stage distribution: IIB (10), III (11), IVA (2); 20 epidermoid and 3 adenocarcinoma; ECOG: 0 (16), 1 (7). 22 pts received the 6 planned cycles. Median dose of irradiation was 45 Gy (43.2–50) with no toxicity related interruption. 17 pts underwent BT, 2 had hysterectomy and 1 received complementary external irradiation 12 Gy. CT dose reduction was necessary in 4 pts (cycle 5 or 6) and cycles postponed for 10 pts (cycle. 5 or 6). One pt experienced paclitaxel allergy at L1. G 3 anemia and/or neutropenia were reported in 11 pts and G 4 neutropenia = 1 week in 2 pts. Radiodermatitis occurred in 5 pts, asthenia in 3 and nausea in 1. One DLT was observed: unendurable G 3 asthenia + G 3 neutropenia and leucopenia at L3. Clinical and radiological complete response was obtained in 13 pts, 5 PRs and 2 SDs in 20 evaluable pts. Conclusions: Acceptable toxicity and optimal irradiation were possible at L4 in 7 pts. These doses are recommended for future phase II studies of concomitant CT/RT in ACC.

[Table: see text]

No significant financial relationships to disclose.