Anne-Sophie Dieudonné

Vitamin D status at breast cancer diagnosis: correlation with tumor characteristics, disease outcome, and genetic determinants of vitamin D insufficiency

We correlated serum 25-hydroxyvitamin D(3) (25OHD) levels with tumor characteristics and clinical disease outcome in breast cancer patients and assessed the impact of genetic determinants of vitamin D insufficiency. We collected serum from 1800 early breast cancer patients at diagnosis, measured 25OHD by radioimmunoassay (RIA), and determined genetic variants in vitamin D-related genes by Sequenom. Multivariable regression models were used to correlate 25OHD levels with tumor characteristics. Cox proportional hazard models were used to assess overall survival (OS), disease-specific survival (DSS), and disease-free interval (DFI). Lower 25OHD serum levels significantly correlated with larger tumor size at diagnosis (P = 0.0063) but not with lymph node invasion, receptor status, or tumor grade. Genetic variants in 25-hydroxylase (CYP2R1) and vitamin D-binding (DBP) protein significantly determined serum 25OHD levels but did not affect the observed association between serum 25OHD and tumor size. High serum 25OHD (>30 ng/mL) at diagnosis significantly correlated with improved OS (P = 0.0101) and DSS (P = 0.0192) and additionally had a modest effect on DFI, which only became apparent after at least 3 years of follow-up. When considering menopausal status, serum 25OHD had a strong impact on breast cancer-specific outcome in postmenopausal patients [hazards ratios for 25OHD >30 ng/mL versus ≤30 ng/mL were 0.15 (P = 0.0097) and 0.43 (P = 0.0172) for DSS and DFI, respectively], whereas no association could be demonstrated in premenopausal patients. In conclusion, high vitamin D levels at early breast cancer diagnosis correlate with lower tumor size and better OS, and improve breast cancer-specific outcome, especially in postmenopausal patients.

Management of menopausal symptoms in breast cancer patients.

In breast cancer patients, menopausal symptoms such as hot flashes, urogenital problems, musculoskeletal symptoms and cognitive dysfunction are common, regardless of age at diagnosis. They affect quality of life and systemic therapy will worsen this. Endocrine and/or chemotherapy may induce temporary or permanent ovarian failure and can exacerbate these symptoms. Hormone therapy (HT) has been studied in breast cancer survivors, but safety has been questioned. The HABITS trial investigating estrogen-based HT, as well as the LIBERATE trial investigating tibolone, found a reduction in disease-free survival for those treated. Alternative strategies are needed, as menopause symptoms may reduce compliance with breast cancer treatments. This article reviews recently published strategies to tackle menopausal problems in breast cancer patients. Antidepressants may help with hot flashes. Acupuncture and hypnosis can also be used but the evidence is conflicting. For urogenital problems vaginal moisturizers or topical estrogens can be employed. A musculoskeletal syndrome induced by aromatase inhibitors (AIs) is frequently encountered and currently there are no effective treatment strategies. Bisphosphonates reduce AI-induced bone resorption and can also increase disease-free and overall survival. Standard-dose endocrine and chemotherapy are associated with a decline in cognitive function.

Genetic variability in the multidrug resistance associated protein-1 (ABCC1/MRP1) predicts hematological toxicity in breast cancer patients receiving (neo-)adjuvant chemotherapy with 5-fluorouracil, epirubicin and cyclophosphamide (FEC)

BACKGROUND To assess the impact of single-nucleotide polymorphisms (SNPs) on predefined severe adverse events in breast cancer (BC) patients receiving (neo-)adjuvant 5-fluorouracil (FU), epirubicin and cyclophosphamide (FEC) chemotherapy. PATIENTS AND METHODS Twenty-six SNPs in 16 genes of interest, including the drug transporter gene ABCC1/MRP1, were selected based on a literature survey. An additional 33 SNPs were selected in these genes, as well as in 12 other genes known to be involved in the metabolism of the studied chemotherapeutics. One thousand and twelve female patients treated between 2000 and 2010 with 3-6 cycles of (neo-)adjuvant FEC were genotyped for these SNPs using Sequenom MassARRAY. Severe adverse events were evaluated through an electronic chart review for febrile neutropenia (FN, primary end point), FN first cycle, prolonged grade 4 or deep (<100/µl) neutropenia, anemia grade 3-4, thrombocytopenia grade 3-4 and non-hematological grade 3-4 events (secondary end points). RESULTS Carriers of the rs4148350 variant T-allele in ABCC1/MRP1 were associated with FN relative to homozygous carriers of the G-allele [P = 0.0006; false discovery rate (FDR) = 0.026]. Strong correlations with secondary end points such as prolonged grade 4 neutropenia (P = 0.002, FDR = 0.046) were also observed. Additionally, two other SNPs in ABCC1/MRP1 (rs45511401 and rs246221) correlated with FN (P = 0.007 and P = 0.01, respectively; FDR = 0.16 and 0.19), as well as two SNPs in UGT2B7 and FGFR4 (P = 0.024 and P = 0.04; FDR = 0.28 and 0.38). CONCLUSION Genetic variability in ABCC1/MRP1 was associated with severe hematological toxicity of FEC.

Tamoxifen metabolism and efficacy in breast cancer- a prospective multicentre trial

Purpose: Levels of endoxifen, the most active metabolite of tamoxifen, vary by the highly polymorphic cytochrome P450 (CYP) 2D6 enzyme. We prospectively investigated tamoxifen efficacy by serum endoxifen levels and the tamoxifen activity score (TAS). Experimental Design: A prospective observational multicenter study included postmenopausal women with an estrogen receptor–positive breast cancer receiving first-line tamoxifen, 20 mg daily in the neoadjuvant or metastatic setting, recruited between February 2009 and May 2014. The primary endpoint was the objective response rate (ORR) using RECIST criteria 1.0. Secondary endpoints were clinical benefit (CB), progression-free survival (PFS), and tolerability of tamoxifen. The main analysis used logistic regression to relate ORR to serum endoxifen levels after 3 months. Endpoints were also related to other tamoxifen metabolites and to TAS. Results: Endoxifen levels were available for 247 of all 297 patients (83%), of which 209 with target lesions (85%). Median follow-up time for PFS was 32.5 months, and 62% progressed. ORR and CB were 45% and 84%, respectively. ORR was not related to endoxifen, and the OR of ORR was 1.008 per μg/L increase in endoxifen (95% confidence interval, 0.971–1.046; P = 0.56). In general, none of the endpoints was associated with endoxifen levels, tamoxifen metabolites, or TAS. Conclusions: Under the prespecified assumptions, the results from this prospective clinical trial do not suggest therapeutic drug monitoring of endoxifen to be of clinical value in postmenopausal women treated with tamoxifen for breast cancer in the neoadjuvant or metastatic setting. Clin Cancer Res; 24(10); 2312–8. ©2018 AACR.

Reproduction Rates After Cytotoxic Therapy

TO THE EDITOR: In the January 20, 2009, issue of Journal of Clinical Oncology, Cvancarova et al described the results of a Norwegian monocentric retrospective analysis of reproduction rates after cytotoxic cancer treatment. Although their findings are interesting, an important shortcoming of their study is that the desire to become pregnant after such therapy was not considered. We believe this is a crucial aspect when looking into reproduction after cytotoxic cancer treatment. We recently sent out a questionnaire, approved by our ethics committee, and report here the results, which contained some complementary aspects on this issue not covered by the Norwegian study. Our data represent relevant complementary information (not included in the Norwegian study) with respect to desire to become pregnant before and after chemotherapy, time to pregnancy interval, time interval to resumption of menstrual cyclicity, and regularity of menstrual cycles after chemotherapy. The main goal of the questionnaire was to assess the effect of cytotoxic therapy on menstrual status, desire to become pregnant, and pregnancy rate in women age 15 to 35 years (highly relevant with respect to desire to become pregnant after chemotherapy) who had been treated for breast cancer or lymphoma (both Hodgkin’s lymphoma [HL] and non-Hodgkin’s lymphoma [NHL]), which represent the most commonly observed cancers in women with fertility problems after chemotherapy. We only considered patients who did not experience relapse until 2007 and identified 126 eligible female patients who had previously received chemotherapy at our institution during the period from 1999 to 2005. The questionnaire together with a patient information form was sent out to all such patients; the forms were completed and sent back for analysis. The response rate was 78% (98 of 126 patients). The mean age of our patients at time of diagnosis was 29 years (range, 15 to 35 years; median, 30 years; standard deviation [SD], 5.6 years), and they were affected by breast cancer (64.3% compared with 11% in the Norwegian study), HL (27.5%), and NHL (8.2%). Chemotherapy administered for breast cancer included cyclophosphamide, methotrexate, and fluorouracil (23.8%); anthracycline-based chemotherapy (61.9%); and an anthracycline plus taxane–based regimen (14.3%). For HL, a regimen of doxorubicin, bleomycin, vinblastine, and dacarbazine was administered, and patients with NHL received cyclophosphamide, doxorubicin, vincristine, and prednisone–like chemotherapy. Overall, our results are in agreement with those of the Norwegian study. Looking at prediagnosis parenthood, 45 (45.9%) of the 98 patients in our study were childless at diagnosis (compared with 34% in the Norwegian study). Fifty three (54.1%) of 98 had been pregnant before, with a mean of 1.7 living births (range, zero to four living births; median, two living births; SD, 0.8 living births). Mean age at first pregnancy before therapy was 25.9 years (range, 18 to 32 years; median, 26 years; SD, 3 years). For patients who had been pregnant before diagnosis, mean number of months between last birth and cancer diagnosis was 38.3 (range, 0 to 132 months; median, 23.5 months; SD, 34.7 months). The pregnancy rate after therapy in our study was slightly less than that in the whole group in the study by Cvancarova et al (9.2% v 13% of female patients). Mean age at diagnosis of our patients who became pregnant after therapy was 27.9 years (range, 16 to 33 years; median, 30 years; SD, 5.5 years). With regard to postdiagnosis reproduction (ie, reproduction rate after cancer diagnosis), we compared the Norwegian results in the 1988 group with those in our group of patients. Two patient groups were defined: patients who were childless at moment of diagnosis and patients who had already had one or more children at diagnosis. Our postdiagnosis reproduction rates were 11% (compared with 19% in the 1988 group) and 7% (compared with 4% in the 1988 group), respectively. In our study, data concerning desire to become pregnant before and after cytotoxic treatment were analyzed. The results of our questionnaire showed that 26 (26.5%) of 98 patients had a desire to become pregnant before therapy, whereas only nine (9.2%) of 98 patients wanted to become pregnant after therapy. In seven (7.1%) of 98 patients who had a desire to become pregnant before therapy, there were fertility problems before diagnosis. After therapy, there were fertility problems in four (44.4%) of the nine patients who wanted to become pregnant. Only four (44.4%) of the nine patients with a clear, active desire to become pregnant actually did so after therapy. One of our patients had a successful pregnancy after therapy despite having not self-reported an active desire to become pregnant. In the report by Cvancarova et al, no information was provided about the desire to become pregnant before or after therapy. In our study, eight (88.9%) of nine patients who became pregnant after therapy had uncomplicated pregnancies (compared with 98% in the Norwegian study), and all births were live births (12 births among eight patients). One patient had an abortion after therapy, in accordance with her wishes. In seven (77.8%) of the nine patients, no major health problems were observed in the newborns. No patient underwent in vitro fertilization treatment in the first postdiagnosis pregnancy (compared with 2% in the Norwegian study). We also obtained information about mean time between start of active desire to become pregnant and successful pregnancy. Before therapy, this was 5.9 months (range, 0 to 48 months; median, 3 months; SD, 9.3 months). After therapy, mean time until successful pregnancy was 10.2 months (range, 0 to 36 months; median, 8 months; SD, 11.5 months). Part of our questionnaire was about menses before, during, and after therapy. During chemotherapy, 34 (34.7%) of 98 patients maintained spontaneous menses without oral contraception. Spontaneous resumption of menstruation was observed 5.5 months (range, 0 to 61 months; median, 9.5 months; SD, 2 months) after cessation of cytotoxic therapy in 68 (69.4%) of 98 patients whose menses had stopped during chemotherapy. In 44 (63.4%) of these 68 patients, menses were regular. Seventy (71.4%) of our 98 patients started contraception after JOURNAL OF CLINICAL ONCOLOGY C O R R E S P O N D E N C E VOLUME 27 NUMBER 27 SEPTEMBER 2

Why Should Results From Metastatic Trials No Longer Matter for Early-Stage Disease?

TO THE EDITOR: The recently published comments by Brauch et al regarding our current knowledge about the importance of CYP2D6 status when prescribing tamoxifen concluded that results from ongoing prospective studies in the metastatic/neoadjuvant setting will not help to close the circle on this topic. We do agree with their conclusion that published retrospective studies—so far—do not provide sufficient evidence to determine whether or not CYP2D6 testing should be performed in daily practice. However, we believe that two registered, ongoing, prospective studies in the neoadjuvant/metastatic setting (the European CYPTAMBRUT-2 study [Study to Assess Response to Tamoxifen in (cT3)/Inoperable Locally Advanced/Metastatic Estrogen Receptor–Positive Breast Cancer by the “Tamoxifen Activity Score” Based on Drug Interaction and Polymorphisms in Genes Coding for Tamoxifen Metabolising Enzymes] and the study of the Eastern Cooperative Oncology Group E3108 [Tamoxifen Citrate in Treating Patients With Metastatic or Recurrent Breast Cancer]) will provide answers that are not available from retrospective studies and which are likely to be relevant in the adjuvant setting. Although metastatic breast cancer differs in many respects from early breast cancer, in which tamoxifen helps to eradicate subclinical disseminated disease, metabolic pathways of tamoxifen for efficacy are disease-stage independent. It is unlikely that endoxifen levels matter in the adjuvant setting if they do not matter in the metastatic setting. However, if endoxifen levels do matter in metastatic disease, it does probably but not necessarily follow that this applies in the adjuvant setting. As principal investigators of the European CYPTAMBRUT-2 study, we would like to stress that this study includes only patients in the first-line metastatic/neoadjuvant setting. Patients with hormoneinsensitive or aromatase inhibitor–refractory disease were excluded from this study, as were premenopausal women. The study has so far recruited 240 of 260 patients. The CYPTAMBRUT-2 study not only addresses response to tamoxifen according to genetic variants but also takes endoxifen levels into account. One should first elucidate the relationship between endoxifen levels and tamoxifen efficacy before investigating the relationship between genetic variants and efficacy. Tamoxifen metabolization is complex; multiple enzymes are involved, and other factors affect endoxifen levels as well. Negative results from studies that have assessed the relationship between genetic variants and efficacy could be a result of the fact that variations in endoxifen levels were not related to efficacy or that the variations in endoxifen levels were not represented well by the analyzed genetic variants. Studies examining the pharmacogenomics of tamoxifen should include the measurement of all relevant metabolites of tamoxifen, as well as all genes and gene polymorphisms that are involved in the metabolism of tamoxifen; in nearly all previous retrospective studies, this has not been the case.

Abstract OT2-1-05: Prospective multicenter study evaluating the effect of impaired tamoxifen metabolization on efficacy in breast cancer patients receiving tamoxifen in the neo-adjuvant or metastatic setting - The CYPTAM-BRUT 2 trial

Background Tamoxifen is commonly used to treat and prevent hormone receptor positive breast cancers. This drug is metabolized into more active metabolites by liver enzymes such as cytochrome P450 (CYP) enzymes. Endoxifen is considered to be the principal active metabolite of tamoxifen. As CYP enzymes are highly polymorphic in humans, endoxifen plasma levels are modulated by the patient’s genotype. It, however, is not yet clear if lowered endoxifen plasma levels have an effect on tamoxifen efficacy. This is the first prospective study where the association between endoxifen plasma concentrations, multiple CYP-genotypes and clinical outcome in postmenopausal patients treated with tamoxifen is investigated. Trial Design CYPTAM-BRUT 2 is a prospective multi-center open label, single-arm, non-randomized observational study. Postmenopausal women with measurable, estrogen receptor positive breast cancer receiving tamoxifen as neo-adjuvant or as first-line metastatic treatment are included in this study. The objective treatment response and clinical benefit are observed to investigate the efficacy of 20 mg tamoxifen daily. Patients are allowed to have started tamoxifen before inclusion but not more than three months. Further, if more than twelve months have passed after completion of the adjuvant therapy prior endocrine therapy in the adjuvant setting is allowed. Patients receiving neo-adjuvant tamoxifen will be assessed no more than four months after starting with tamoxifen. The primary endpoint is a statistical association between steady-state endoxifen plasma concentrations and the objective response rate (ORR) after 3-6 months of tamoxifen, under the assumption that the relationship is linear with an odds ratio (OR) of 1.49 per 10 nmol/L. Using available data on endoxifen concentrations, this OR is chosen to reflect an improvement from 10% ORR in the lowest endoxifen quartile to 30% in the highest endoxifen quartile when the overall ORR is around 18%. To have 90% power at a 5% significance level, 180 patients have to be included into the study. The main secondary study endpoint is the relation between endoxifen plasma concentrations and clinical benefit (CR+PR+SD at 6 months). The study has to include 270 patients to detect a statistically significant association with endoxifen with 90% power at a 5% significance level, assuming an OR of 1.28 per 10 nmol/L. This OR is chosen to reflect an improvement of clinical benefit at 6 months from 30% in the lowest endoxifen quartile to 50% in the highest endoxifen quartile (overall clinical benefit around 39%). For both endpoints the RECIST criteria are used. Other endpoints are progression-free survival, tolerability of tamoxifen treatment and the association between CYP2D6 genotype and clinical outcome. Patient accrualPatients from 22 participating centers in Belgium and Switzerland are included in this trial. In May 2014, the predefined sample size of 270 patients was reached. Follow-up of the last patients will continue until all required data are obtained (i.e blood samples and response evaluation). Citation Format: Kathleen Van Asten, Lynn Jongen, Anne-Sophie Dieudonne, Anneleen Lintermans, Chantal Blomme, Olivier Brouckaert, Diether Lambrechts, Hans Wildiers, Marie-Rose Christiaens, Dirk Timmerman, Ben Van Calster, Jan Decloedt, Patrick Berteloot, Didier Verhoeven, Markus Joerger, Khalil Zaman, Vincent Dezentje, Patrick Neven. Prospective multicenter study evaluating the effect of impaired tamoxifen metabolization on efficacy in breast cancer patients receiving tamoxifen in the neo-adjuvant or metastatic setting - The CYPTAM-BRUT 2 trial [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr OT2-1-05.