P2-17-09: A Prospective Assessment of Loss of Grip Strength by Baseline BMI in Breast Cancer Patients Receiving Adjuvant Aromatase Inhibitors or Tamoxifen

Background The 3rd generation aromatase inhibitors (AIs) induce or enhance musculoskeletal problems. Underlying mechanisms are probably multiple, but remain unknown. We have previously reported that loss of grip strength together with tenosynovial abnormalities are more important in AI- than in tamoxifen-users (Morales et al, JCO 2008) and that musculoskeletal changes in AI-users are more pronounced in women with extremes in baseline BMI (Lintermans et al, Ann Oncol 2011) We here report preliminary results from a larger population and plan to validate findings in patients from University of Michigan.

Patients and methods In this prospective observational study, postmenopausal early breast cancer patients scheduled to start adjuvant hormonal therapy with any of the third generation AIs or tamoxifen were recruited. After providing informed consent, a functional assessment test of grip strength was performed with a modified sphygmomanometer. Re-evaluation was done after 3, 6 and 12 months of therapy. BMI and waist to hip ratio (WHR) were assessed and a rheumatological questionnaire was completed at each visit.

Results Ninety-four (79 AI; 15 tamoxifen) of the planned 200 patients were included in this preliminary study. 18% of AI-users discontinued their treatment due to musculoskeletal symptoms compared with none of the tamoxifen-users. 62% of patients on AI and 35% of tamoxifen patients complained of new or worsened joint pain. Table 1 shows the proportion of patients that attribute their complaints to their endocrine therapy. Grip strength significantly decreased over time (p=0.03), with patients under AI treatment having a larger loss of grip strength than patients under tamoxifen treatment (p=0.04). We confirm our previously reported results on the shape of the curve “BMI AI-induced loss of grip strength”. Years past menopause and age showed a significant effect on grip strength (p<0.001 and p=0.0007, respectively), whereas no significant relationship was found between WHR with grip strength or joint pain. Detailed results will be presented.

Ineffective attempt to preserve fertility with a levonorgestrel-releasing intrauterine device in a young woman with endometrioid endometrial carcinoma: a case report and review of the literature

BACKGROUND

The treatment of endometrial cancer in young women who want to preserve their fertility is challenging.

CASE

A 25-year-old woman (A0P0G0) was diagnosed with grade 1 endometrioid endometrial carcinoma (EEC). Imaging studies including transvaginal ultrasound (TVS), computed tomography and magnetic resonance imaging (MRI) could not detect myometrial invasion or metastatic disease. The immunohistochemical expression of the estrogen and progesterone receptor in the tumor was strongly positive, whereas p53 staining was negative. After extensive counseling, we decided to use a levonorgestrel-releasing intrauterine device to preserve her fertility. Follow-up was organized every three months and consisted of serum CA125 levels, TVS, endometrial biopsy and MRI. The tumor regressed after ten months and the intrauterine device was removed. However, nine months later, recurrent EEC was diagnosed and a hysterectomy performed. Pathological examination confirmed Stage Ia EEC.

CONCLUSION

Despite the presence of favorable prognostic factors of EEC as determined by grade and immunohistochemistry, the levonorgestrel-releasing intrauterine device was unable to preserve fertility.

Abstract P4-08-09: The Prognostic Importance of “Detection Mode” and “Palpability” in Primary Operable Grade 2 Breast Cancers

Background: Histologic grade 2 breast cancer may benefit most from molecular classification for prognosis. We studied the prognostic value of ‘detection mode’ and ‘palpability’ together with other prognostic variables in a consecutive series of grade 2 lesions.

Patients and methods: We used follow-up data from new diagnosed and primary operable grade 2 breast cancers from UH Leuven (January 2000 - May 2005). Endocrine, chemo-and radiotherapy were given when indicated. The first appearing breast cancer related event (BCRE) was calculated as local (local, regional or contra-lateral) or metastatic (metastatic if both appeared together). These prognostic variables were assessed: age at diagnosis (per decade increase), tumor size (per cm increase), screening (yes/no), palpability (yes/no), lobular type (yes/no), chemotherapy (yes/no), endocrine therapy (yes/no), lymph node status (0, 1, 2-4, >4), combined expression of IHC-measured ER (pos = any expression), PgR (pos if any expression) and HER-2 (positive = IHC2/3+ and FISH pos), architectural, nuclear and mitotic score. Univariable and multivariable proportional hazards Cox regression using the Firth's penalization method to reduce bias in the parameter estimates were used.

Results: Of the 1013 patients in the dataset, 14 had missing values (1.4%). With a median follow-up of 80 months, we observed 116 BCRE (11.6%); 43 local and 73 metastatic. In univariable analysis, screen detected cancers showed better (HR 0.63, 95% CI 0.42-0.93) and palpable cancers worse prognosis (HR 1.96, 95% CI 1.13-3.71) than non-screen detected and non-palpable cases respectively. The beneficial univariable effect of screening weakened in the multivariable analysis (HR 0.83, 95% CI 0.52-1.31) as 582/602 (97%) of non-screen detected cancers were palpable vs 227/397 (57%) for screen-detected cancers. If palpable was omitted from the multivariable model the HR of screening decreased to 0.68 (95% CI 0.44-1.03) thereby approximating its univariable result. Conclusion: Overall, palpability is more strongly related to prognosis, but detection method also has an effect even though we do not have enough events to show this with a convincing level of reliability.