Search

PURPOSE: Enterocolitis is among the leading adverse events associated with immune checkpoint inhibitors (ICIs). There are limited retrospective data regarding the safety of ICIs in patients with inflammatory bowel disease (IBD; ulcerative colitis, Crohn’s disease) because they have been generally excluded from clinical trials testing ICIs. Furthermore, there are no outcome data available in patients with microscopic colitis, a leading cause of chronic diarrhea. We aimed to study the safety of ICIs in patients with cancer with pre-existing IBD or microscopic colitis. METHODS: We retrospectively reviewed the records of patients with cancer treated at our institution who received at least 1 dose of either a programmed cell death-1 (PD-1)/ PD-1 ligand inhibitor, cytotoxic T-lymphocyte-associated antigen 4 inhibitor, or both between 2011 and 2018. We identified patients with pre-existing IBD or microscopic colitis. RESULTS: Of 548 patients with solid tumor treated with an ICI, we identified 25 with pre-existing colitis (21 IBD; 4 microscopic colitis). An enterocolitis flare occurred in 7 patients (28%): 3 of 4 patients (75%) with microscopic colitis and 4 of 21 (19%) with IBD. All were treated with systemic corticosteroids, 2 required an anti–tumor necrosis factor agent, and one required an anti-integrin agent and colectomy for treatment of refractory colitis. ICI therapy was discontinued in all patients who experienced an enterocolitis flare. CONCLUSION: In our cohort, exacerbation of enterocolitis occurred in a notable percentage of patients with IBD and a majority of patients with microscopic colitis, leading to discontinuation of ICIs. Although these data suggest that patients with cancer with pre-existing IBD/microscopic colitis may be treated with ICIs, additional studies are needed to validate our results.

e18781 Background: Immunotherapy can boost a patient’s immune system to bring about anti-tumor activity and may theoretically exaggerate the inflammatory immune response to infection or vaccinations. Conversely, vaccination may enhance immune activity in patients on immunotherapy. Currently, there are limited safety data in patients treated with immunotherapy who received COVID-19 vaccines. Methods: An IRB-approved retrospective review of Dana-Farber Cancer Institute patients treated with PD-1 checkpoint inhibitors and received COVID-19 vaccines was performed. The primary endpoint was the incidence rate of immune-related adverse events (irAEs) pre- and post- COVID-19 vaccinations. An Electronic Health Record search identified all patients who completed COVID-19 vaccines between 12/1/2020 and 3/31/2021 and received anti-PD-1 monotherapy between 10/1/2020 and 5/31/2021. Medical record review identified timing, occurrence, and grading of irAEs as defined by clinicians or Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. Results: A total of 229 patients were included in the study. The most common underlying cancers were lung (32%) and melanoma (26%). The majority of patients had Pfizer-BioNTech COVID-19 vaccine; 64% of all patients received the third dose of vaccine. The median duration of treatment was 15 months, and median duration of follow-up was 19 months. Among the study cohort, 120 irAEs occurred before vaccination (any grade); 30 (25%) were grade 1, 35 (29%) were grade 2, 11 (9%) were grade 3 or 4, and 44 (37%) were ungraded. The main types of irAEs were skin toxicities (37% of all irAEs) followed by thyroid dysfunction (22%) and colitis (11%). Of the 75 irAEs that occurred after vaccination, 24 (32%) were grade 1, 16 (21%) were grade 2, 16 (21%) were grade 3 or 4, and 19 (26%) were ungraded. The main irAEs types were skin toxicity (31% of all irAEs) followed by thyroid dysfunction (13%) and hepatic toxicity (12 %). The all-grade pre- and post-vaccine irAE incidence rates per 100 patient-months were 4.13 and 5.09, respectively. Within this cohort, there were 0.96 (95% CI -0.362-2.284, p = 0.15) more irAEs post vaccine per 100 patient-months of anti-PD-1 therapy. Conclusions: The data weakly suggested that there is a higher incidence of irAEs post COVID-19 vaccine; however, the comparison did not achieve statistical significance. Further analysis of other patient characteristics will help to determine if any specific patterns are observed.[Table: see text]

2030 Background: Glioblastoma (GBM) and tumor endothelial cells express phosphatidylserine (PS), a highly immunosuppressive membrane phospholipid. PS receptors engage with immune cells, leading to expansion of myeloid-derived suppressor cells (MDSCs) which promote an immunosuppressive and pro-angiogenic tumor microenvironment. Bavituximab (BAV) – a chimeric monoclonal antibody – binds to β2-glycoprotein 1 (β2-GP1) to form a complex of β2-GP1 with PS, resulting in immune activation against tumor cells and anti-angiogenic effects. Pre-clinical data in GBM models suggest synergistic effects of PS blockade, radiation (RT), and temozolomide (TMZ). Here, we present results from a phase II trial (NCT03139916) of BAV, RT and TMZ in GBM patients. Methods: 33 adults with newly diagnosed IDH-wild-type GBM were enrolled and received 6 weeks of RT+TMZ, followed by 6 cycles of TMZ. BAV (3 mg/kg) was given weekly, starting at week 1 of RT+TMZ, for 18 weeks with the option to continue if tolerated. The primary endpoint was the proportion of patients alive at 12 months (OS-12). The null hypothesis would be rejected if OS-12 was ≥ 72%. As an exploratory endpoint, the immune profile in tumor tissue and peripheral blood mononuclear cells (PBMCs) was assessed using nanoString and multispectral immunofluorescence, with the goal to assess on-target effects of BAV in longer vs. shorter surviving patients (split based on median survival). Relative cerebral blood flow (rCBF) from dynamic susceptibility contrast MRI was also obtained. Results: 24 patients were alive at 12 months and OS-12 was 73% (95% CI 59-90%) so the study met its primary endpoint. Median OS was 15.4 months. As best response, 79% of patients had stable disease, 12% had a partial response and 9% had progressive disease. Eight grade 3 or 4 adverse events were seen (no grade 5 AEs). Ten pre-treatment and 7 post-treatment tissue samples were available. Analysis of RNA from pre-treatment tumor specimens showed a significantly positive shift in myeloid-related gene expression in patients with longer survival, with enrichment of 116 and 120 transcripts as well as downregulation of 2 and 1 gene for PFS and OS, respectively. There was no differential expression in PBMCs. Including all tissue samples, there was a marked reduction of MDSCs after BAV compared to time of diagnosis (p = 0.011). Decreased rCBF post-RT/pre-cycle 1 TMZ was associated with improved OS (HR 4.63, p = 0.029). Conclusions: OS-12 was 73%, meeting the primary endpoint and suggesting potential activity of BAV in newly diagnosed GBM. BAV leads to on-target depletion of intratumoral immunosuppressive MDSCs and anti-angiogenic effects. As expected, based on the mechanism of action of BAV, there was no difference in PBMC gene expression profile in patients with long and short survival. Combining BAV with immune checkpoint inhibitors in the future may augment tumor immune response. Clinical trial information: NCT03139916.

12124 Background: Hot flashes are a common side effect of endocrine therapy in breast cancer patients (pts). In a coordinated multinational study, we evaluated the impact of acupuncture on hot flashes and related symptoms in hormone receptor-positive breast cancer pts undergoing adjuvant endocrine therapy in the USA, China, and South Korea. Methods: Parallel randomized trials were conducted at three sites: Dana-Farber Cancer Institute, USA; Jiangsu Hospital of Traditional Chinese Medicine, China; and Daegu Catholic University Medical Center, Republic of Korea, using the same inclusion/exclusion criteria, randomization, and measures. Breast cancer pts receiving adjuvant endocrine therapy and having ≥14 hot flashes per week were randomized equally to acupuncture (A) or usual care (UC); randomization was stratified by number of hot flashes per day (2-6 or 7+). Pts in arm A received a standardized acupuncture protocol twice per week for 10 weeks; pts in the UC arm received usual care. Symptoms were assessed at baseline and week 10. The primary endpoint was change in the Endocrine Symptom Subscale (ESS) of the Functional Assessment of Cancer Therapy-Endocrine Systems (FACT-ES). Secondary endpoints included changes in the FACT-Breast (FACT-B) and Daily Hot Flash Score (DHFS), a measure incorporating hot flash frequency and severity. Target enrollment was 160 pts (USA 80, China 40, Korea 40) providing 84% power to detect a clinically meaningful change in the ESS. Analyses were intention-to-treat. A pooled analysis of individual patient data was performed using linear models adjusted for site and baseline number of hot flashes. Results: Between Jan. 2019 and June 2021, 158 female pts with stage 0-III breast cancer were randomized from the USA (n=78), China (n=40), and South Korea (n=40). Median age was 48 years (range: 25 to 73). At baseline, pts reported similar numbers of hot flashes between study arms (6.2± 4.3 vs. 6.5± 3.8 per day). At week 10, pts in arm A reported statistically significant improvements in ESS score, FACT-B total score and DHFS compared with the UC arm (Table). The reduction in the DHFS from baseline in arm A was 53%. There were no serious adverse events. Conclusions: Acupuncture led to statistically and clinically meaningful improvements in hot flashes, endocrine symptoms, and breast cancer-specific quality of life in women undergoing adjuvant hormonal therapy for breast cancer in the USA, China and South Korea. Clinical trial information: NCT00797732, ChiCTR2100045888, KCT0003618. [Table: see text]

Purpose Cholesterol-lowering medication (CLM) has been reported to have a role in preventing breast cancer recurrence. CLM may attenuate signaling through the estrogen receptor by reducing levels of the estrogenic cholesterol metabolite 27-hydroxycholesterol. The impact of endocrine treatment on cholesterol levels and hypercholesterolemia per se may counteract the intended effect of aromatase inhibitors. Patients and Methods The Breast International Group (BIG) conducted a randomized, phase III, double-blind trial, BIG 1-98, which enrolled 8,010 postmenopausal women with early-stage, hormone receptor–positive invasive breast cancer from 1998 to 2003. Systemic levels of total cholesterol and use of CLM were measured at study entry and every 6 months up to 5.5 years. Cumulative incidence functions were used to describe the initiation of CLM in the presence of competing risks. Marginal structural Cox proportional hazards modeling investigated the relationships between initiation of CLM during endocrine therapy and outcome. Three time-to-event end points were considered: disease-free-survival, breast cancer–free interval, and distant recurrence–free interval. Results Cholesterol levels were reduced during tamoxifen therapy. Of 789 patients who initiated CLM during endocrine therapy, the majority came from the letrozole monotherapy arm (n = 318), followed by sequential tamoxifen-letrozole (n = 189), letrozole-tamoxifen (n = 176), and tamoxifen monotherapy (n = 106). Initiation of CLM during endocrine therapy was related to improved disease-free-survival (hazard ratio [HR], 0.79; 95% CI, 0.66 to 0.95; P = .01), breast cancer–free interval (HR, 0.76; 95% CI, 0.60 to 0.97; P = .02), and distant recurrence–free interval (HR, 0.74; 95% CI, 0.56 to 0.97; P = .03). Conclusion Cholesterol-lowering medication during adjuvant endocrine therapy may have a role in preventing breast cancer recurrence in hormone receptor–positive early-stage breast cancer. We recommend that these observational results be addressed in prospective randomized trials.

3059 Background: Immune checkpoint inhibitors (ICI) have recently been shown to be effective for brain metastases (BM) in melanoma and lung cancer. Several studies demonstrate that 20-50% of BM patients respond to ICI. The reasons behind this wide variability in treatment response is not clear. Therefore, using physiologic imaging, we seek to identify the longitudinal biological changes exerted on BM as a result of ICI administration. Methods: Given the importance of aberrant tumor vasculature in cancer proliferation, we have focused on assessing changes in vascular physiology. We analyzed standard post-contrast and dynamic susceptibility contrast (DSC) MRI to identify characteristic vascular signatures as part of an ongoing Phase 2 study of pembrolizumab for patients with untreated or progressive, previously treated BM from any histology. Tumor volume measurements were calculated by summating all enhancing voxels. As per modified RECIST and RANO criteria for immunotherapy, volumetric increase of > 40% was defined as progressive disease (PD), a decrease of > 60% as partial response (PR), and stable disease (SD) as between -60% and +40%. Results: 35 patients, out of the total cohort of 60, have undergone DSC-MRI analysis. Histologies include 15 with breast cancer, 6 with non-small cell lung cancer, 4 with melanoma, and 10 with other cancers. At baseline, the total number of BM was 1-50+ per patient. Based on summing the entire enhancing intracranial disease burden, best volumetric responses for the 35 evaluable patients include 4 PR, 12 SD, and 19 PD. Thus far, we found that ICI-resistant BM had a 50% increase in cerebral blood flow (CBF), 105% increase in cerebral blood volume (CBV), a 15% increase in mean transit time (MTT), and an 80% increase in vessel caliber at 6 weeks post-treatment. On the other hand, ICI-responsive BM had no change in CBF, a 33% increase in CBV, a 10% decrease in MTT, and no change in vessel caliber. Ongoing analysis to uncover additional vascular changes (e.g. tumor oxygenation, vessel size index) within BM to ICI are pending. Conclusions: Our data provides evidence that effective ICI for BM is associated with unique intra-tumoral vascular physiology. With final analysis, we will uncover other facets of vascular physiology that correlate with ICI response, and may reveal mechanisms of response/resistance within tumors to ICI.

2529 Background: Immune checkpoint inhibitors (ICI) have recently been shown to be effective for brain metastases (BM) for melanoma and lung cancer. This breakthrough has prompted interest in evaluating ICI in BM of other histologies. However, accurately assessing intracranial response in patients undergoing ICI is a challenge, as current measures cannot distinguish pseudoprogression from true tumor progression. To shed light on potential biomarkers of response, we prospectively use perfusion MRI to identify characteristic vascular signatures in a BM-specific trial of ICI. Methods: As part of an ongoing phase II study of pembrolizumab for patients with untreated or progressive, previously treated BM from any histology, patients underwent advanced MRI that includes tumor volume measurements and perfusion imaging with dynamic susceptibility contrast MRI. To calculate volumetric radiographic response, all enhancing voxels were summated. A volumetric increase of >40% was categorized as progressive disease (PD), a decrease of >60% as partial response (PR), and stable disease (SD) as between -60% and +40%. Results: 53 patients have been enrolled, of whom 44 have received at least baseline advanced MR imaging. Histologies include 21 with breast cancer, 5 with non-small cell lung cancer, 4 with melanoma, and 13 with other cancers. At baseline, the total number of BM was 1-50+ per patient. Based on summing the entire enhancing intracranial disease burden, best volumetric responses for the 33 evaluable patients include 4 PR, 10 SD, and 19 PD. On preliminary analysis, there was a correlation between increased tumor cerebral blood volume/flow with tumor progression. Correlation of additional vascular physiologic parameters (e.g. vessel caliber, tissue oxygenation) and volumetric response to patient outcome and standardized response criteria (iRANO) are ongoing. Conclusions: Pembrolizumab likely has anti-tumor efficacy in BM. Our data provides potential evidence that effective ICI is associated with a decrease in perfusion. Ongoing analyses to uncover additional vascular changes – specifically longitudinal metrics reflecting vascular structure and function - within BM to ICI are pending. These findings have potential to illustrate mechanisms of efficacy for ICI and biomarkers of response in this patient population.

3609 Background: Combination BRAF and MEK inhibitor therapy is associated with response in patients (pts) with BRAF mutant (mut) solid tumors; however critical limitations for the durable activity of these agents remains. Preclinically, the addition of heat shock protein 90 (HSP90) inhibitors improves the efficacy of BRAF inhibitor (BRAFi) therapy in both BRAFi -sensitive and resistant mutant cell lines. Methods: CTEP study 9557 (NCT02097225) is a phase I study designed to determine the safety and efficacy of the small molecule HSP90inhibitor, AT13387, in combination with dabrafenib (dab) and trametinib (tram) in patients with BRAF V600E/K mut solid tumors. Prior chemotherapy, immunotherapy, BRAF and/or MEK exposure was permitted. The primary objective was to determine the maximum tolerated dose (MTD). Results: From July 2015 to June 2018, 22 patients with previously treated, metastatic BRAF V600E/K mut solid tumors were enrolled using a 3 + 3 design at four dose levels (DL) (Table). Pts were predominantly female (59%) with a median age of 57.5yrs (37 -75). The most common tumor type was BRAF V600Emut colon cancer (N=12). Dose limiting toxicities (DLTs) occurred in one patient in DL3 and one in DL4, specifically grade 3 myelosuppression and fatigue, respectively. The MTD was Dab 150mg [BID/PO], Tram 2mg [QD/PO] and AT1187 260mg/m2 [D1,8,15/IV]. Twenty-one of 22 pts were eligible for efficacy assessment. Best response, per RECIST 1.1, was partial response (PR) in 2 pts – one with colon ca (TKI-naïve), one with melanoma (TKI-resistant) - stable disease (SD) in 8 pts, and disease progression (PD) in 11 with a disease control rate (PR + SD) of 47.6% (90% CI: 29% - 67%). Median time to progression was significantly longer in DL3 (3.9 mths; 1.8-9.2) compared to DL1 (1.6mths; 0.9-1.7) or DL2 (1.5; 0.6-3.6). Median PFS and OS were 1.8mths (90% CI: 1.6 – 3.7mths) and 5.1 mths (90% CI: 2.5 -10.6mths), respectively. Median OS was not reached in DL3/4. Correlative data on the expression of the key signaling proteins relating to response will be presented at the meeting. Conclusions: HSP90 inhibition combined with BRAF/MEK inhibition was determined to be safe with evidence of disease control in a heavily pre-treated population of pts with BRAF V600E/K mut solid tumors. Clinical trial information: NCT02097225 . [Table: see text]

10036 Background: Uveal melanoma is a rare and aggressive subset of melanoma that is minimally responsive to traditional therapies. Greater than 80% of uveal melanomas have a mutation in GNAQ or GNA11 which lead to downstream signaling through the MAPK pathway. This has led to efforts to treat uveal melanoma with MEK inhibition with mixed results. Ulixertinib (BVD-523) is a potent and reversible small molecule ATP-competitive inhibitor of both ERK1 and ERK2 protein kinases which has undergone phase I testing. Methods: We performed a phase II study to determine the efficacy and safety of BVD-523 in patients with metastatic uveal melanoma. This was conducted as a Simon two-stage design with a total sample size of 25 patients (pts) and an initial evaluation of efficacy after 13 pts. Two responses were required to continue to the second stage. Results: From April 2018 to April 2019 thirteen pts were enrolled. Pts were predominantly female (69%) with a median age of 64 yrs. (34 -76). Sites of metastasis included liver (84.6%) and lung (30.8%). Grade 3 and 4 toxicities associated with therapy were consistent with BVD-523 and other ERK inhibitors and included LFT elevation, hyponatremia, pruritis, amylase elevation, anemia and rash. The best response, per RECIST 1.1, was stable disease (SD) in 4 pts, and disease progression (PD) in 7 patients. Two patients were unevaluable for response due to withdrawing themselves from the study. Median time to progression was 2.0 months (90% CI: 1.8 – 3.6 mos.). There were eight deaths due to disease progression with a median survival time of 6.9 months (90%CI: 3.2 to 8.3 mos.). Analysis of correlative data from pre- and on-treatment biopsies exploring the change in expression of key signaling proteins relating to treatment is underway. Conclusions: ERK inhibition with ulixertinib (BVD-523) did not demonstrate activity in patients with metastatic uveal melanoma. The toxicities observed on study were consistent with what would be expected with MAPK pathway inhibition. Clinical trial information: NCT03417739.

89 Background: There is a lack of predictive markers informing on the risk of colitis in patients treated with immune checkpoint inhibitors (ICIs). The aim of this study was to identify potential factors associated with development of ICI colitis. Methods: We performed a retrospective analysis of melanoma patients at Dana-Farber Cancer Institute who received PD-1, CTLA-4, or combination blockade between May 2011 to October 2017. Clinical and laboratory characteristics associated with pathologically confirmed ICI colitis were evaluated using multivariate logistic regression analyses. External validation was performed on an independent cohort from Massachusetts General Hospital. Results: The discovery cohort included 213 patients of whom 37 developed ICI colitis (17%). The odds of colitis were higher in patients treated with ipilimumab either as monotherapy or in combination with nivolumab compared to those treated with pembrolizumab. Vitamin D use was recorded in 66/213 patients (31%) before starting ICIs. In multivariable regression analysis, vitamin D use conferred significantly reduced odds of developing ICI colitis (OR 0.35, 95% CI 0.1–0.9). These results were confirmed in the validation cohort of 169 patients of whom 49 developed ICI colitis (29%). Pretreatment neutrophil/lymphocyte ratio (NLR) ≥5 predicted reduced odds of colitis (OR 0.34, 95% CI 0.1–0.9) only in the discovery cohort. Conclusions: This is the first study to report that among patients treated with ICIs, vitamin D intake is associated with reduced risk for ICI colitis. This finding is consistent with prior reports of prophylactic use of vitamin D in ulcerative colitis and GVHD. This observation should be validated prospectively in future studies. [Table: see text]

155 Background: Ang-2 is produced by endothelial cells, predominantly in tissues undergoing vascular remodeling. Current studies suggest that Ang-2 partially suppresses T cell activation by increasing PD-L1 expression and decreasing activation of monocytes. We demonstrated that high pre-treatment serum Ang-2 is associated with reduced overall survival in patients treated with PD-1 blockade. We therefore hypothesized that the combination of Ang-2 and PD-1 blockade may be effective for treatment of patients with advanced cancer. Methods: We initiated a phase 1b trial of the combination of pembrolizumab and trebananib, an Ang-1/2 neutralizing peptibody in advanced solid tumors. Treatment consisted of an induction phase of pembrolizumab 200 mg every 3 weeks and trebananib weekly (with an initial run-in dose escalation of 15-30 mg/kg) for 12 weeks followed by pembrolizumab alone for 2 years. Here we present the updated data in the fully enrolled CRC expansion cohort. Results: The study enrolled 18 microsatellite stable (MSS) CRC patients. There were no DLTs in the dose escalation phase, and 30mg/kg was deemed to be the MTD. This summary is based on 15 CRC patients treated with 30 mg/kg trebananib plus pembrolizumab. The most common treatment-related adverse events (AEs) were abdominal distension, diarrhea, limb edema, transaminitis, and proteinuria, each reported in 40% of the patients. As of September 2019 (median follow up of 10 months), 13 patients were off treatment and two were continuing. Eleven patients were off treatment due to disease progression and 2 due to unacceptable toxicities (grade 4 pneumonitis and grade 3 transaminitis). The response rate was 7% (1 partial response for 23.8+ months) and the disease control rate was 27%, with 4 stable disease for a median of 10 months (4-18 months). Median time to progression and overall survival were 2.8 months (90% CI: 1.5 to 8.1 months) and 9.0 months (90% CI: 2.6 months to ∞), respectively. Conclusions: The combination of pembrolizumab and trebananib is well tolerated and demonstrated promising activity in patients with heavily treated MSS CRC. Clinical trial information: NCT03239145.

Purpose To evaluate the relative effectiveness of letrozole compared with tamoxifen for patients with invasive ductal or lobular carcinoma. Patients and Methods Patients diagnosed with early-stage invasive ductal carcinoma (IDC) or classic invasive lobular carcinoma (ILC) who were randomly assigned onto the Breast International Group (BIG) 1-98 trial and who had centrally reviewed pathology data were included (N = 2,923). HER2-negative IDC and ILC were additionally classified as hormone receptor–positive with high (luminal B [LB] –like) or low (luminal A [LA] –like) proliferative activity by Ki-67 labeling index. Survival analyses were performed with weighted Cox models that used inverse probability of censoring weighted modeling. Results The median follow-up time was 8.1 years. In multivariable models for disease-free survival (DFS), significant interactions between treatment and histology (ILC or IDC; P = .006) and treatment and subgroup (LB like or LA like; P = .01) were observed. In the ILC subset, there was a 66% reduction in the hazard of a DFS event with letrozole for LB (hazard ratio [HR], 0.34; 95% CI, 0.21 to 0.55) and a 50% reduction for LA subtypes (HR, 0.50; 95% CI, 0.32 to 0.78). In the IDC subset, there was a significant 35% reduction in the hazard of a DFS event with letrozole for the LB subtype (HR, 0.65; 95% CI, 0.53 to 0.79), but no difference between treatments was noted for IDC and the LA subtype (HR, 0.95; 95% CI, 0.76 to 1.20). Conclusion The magnitude of benefit of adjuvant letrozole is greater for patients diagnosed with lobular carcinoma versus ductal carcinoma.

e14160 Background: Ang-2 is produced by endothelial cells, predominantly in tissues undergoing vascular remodeling and in hypoxic microenvironments. Ang-2 partially suppresses T cell activation by increasing PD-L1 expression and decreasing monocyte activation. Our group demonstrated that high pre-treatment serum Ang-2 is associated with reduced overall survival (OS) in patients treated with PD-1 blockade. Methods: We initiated a phase 1b trial to evaluate the safety, efficacy, and immunological effect of the combination of pembrolizumab and trebananib, an Ang-1/2 neutralizing peptibody, in solid tumors. Treatment consisted of an induction phase of pembrolizumab 200 mg every 3 weeks and trebananib weekly (with an initial run-in dose escalation of 15-30 mg/kg) for 12 weeks followed by pembrolizumab alone for 2 years. Here we present the preliminary data in the fully enrolled CRC expansion cohort. Results: The study enrolled a total of 18 microsatellite stable (MSS) heavily pretreated CRC patients. There were no DLTs and 30mg/kg was deemed to be the MTD. This summary is based on 15 CRC patients treated with 30 mg/kg trebananib (3 in dose escalation, 12 in dose expansion) plus pembrolizumab. Nine patients (60%) were female and the median age was 51 years (43-63). The most common treatment-related adverse events (AEs) were abdominal distension, diarrhea, limb edema, transaminitis, and proteinuria (each reported in 40% of the patients). One grade 3 transaminitis and one grade 4 pneumonitis related to pembrolizumab were reported. As of the date of data retrieval, 11 patients were off treatment and four were continuing with a median follow-up of 6.1 months. The response rate was 7% (1 partial response for 14 months) and the disease control rate was 33% (4 stable disease with CEA trending down). Median time to progression and OS were 2.8 (90% CI: 1.5 to 4.9 months) and 11.4 months (90% CI: 2.6 months to ∞), respectively. Conclusions: The combination of pembrolizumab and trebananib is well tolerated and demonstrated promising activity in heavily pretreated MSS CRC. Paired biopsies and blood were collected for planned correlative immune analyses which will be presented at the meeting. Clinical trial information: NCT03239145.

49 Background: Pre-clinical and clinical evidence support immune effects of targeted radiation (RT) resulting in synergy with immune checkpoint therapy. However, prospective data are lacking in regards to safety, efficacy and immunologic impacts of different doses of RT given in conjunction with PD-L1 and CTLA-4 blockade. Methods: ETCTN 10021 is a multicenter randomized phase 2 study evaluating repeated low-dose fractionated RT (LDFRT) or hypofractionated RT (HFRT) in combination with PD-L1 and CTLA-4 inhibition in metastatic microsatellite-stable colorectal cancer (CRC) refractory to first-line chemotherapy, or non-small cell lung cancer progressive on prior PD-1 inhibition. The primary endpoint is objective response excluding irradiated lesions. Pre- and on-treatment biopsies were analyzed using multiplex immunofluorescence (IF) evaluating multiple immune markers. We report the prespecified analysis of the CRC cohort. Results: Twenty CRC patients were randomized, 19 of whom were treated with immunotherapy and RT, and 18 of whom were evaluable for response. Median lines of prior therapy were 4 (range 1-7). There were 16 patients with toxicity potentially related to treatment (84%), and 8 patients with possibly associated grade 3-5 toxicity (42%). Best response was stable disease in one patient with an abscopal out-of-field response who developed new lesions after HFRT and 4 cycles with decreasing overall disease burden. Correlative IF performed on 5 sample pairs revealed that HFRT, but not LDFRT, increased infiltration of CD8+, and CD8+/PD1+/Ki67+ T-cells in the RT field. Conclusions: We demonstrate the feasibility of a multicenter randomized study adding different RT regimens to PD-L1/CTLA-4 blockade. Toxicity was, in general, consistent with PD-L1/CTLA-4 inhibition with no significant RT-related toxicities noted. Although the best response of stable disease doesn’t support the use of PD-L1/CTLA-4 inhibition with HFRT or LDFRT in this population using these RT regimens and schedule, integrated biomarkers provide support that HFRT impacts the local immune microenvironment. Clinical trial information: NCT02888743.

Purpose Amplifications and mutations in the KIT proto-oncogene in subsets of melanomas provide therapeutic opportunities. Patients and Methods We conducted a multicenter phase II trial of imatinib in metastatic mucosal, acral, or chronically sun-damaged (CSD) melanoma with KIT amplifications and/or mutations. Patients received imatinib 400 mg once per day or 400 mg twice per day if there was no initial response. Dose reductions were permitted for treatment-related toxicities. Additional oncogene mutation screening was performed by mass spectroscopy. Results Twenty-five patients were enrolled (24 evaluable). Eight patients (33%) had tumors with KIT mutations, 11 (46%) with KIT amplifications, and five (21%) with both. Median follow-up was 10.6 months (range, 3.7 to 27.1 months). Best overall response rate (BORR) was 29% (21% excluding nonconfirmed responses) with a two-stage 95% CI of 13% to 51%. BORR was significantly greater than the hypothesized null of 5% and statistically significantly different by mutation status (7 of 13 or 54% KIT mutated v 0% KIT amplified only). There were no statistical differences in rates of progression or survival by mutation status or by melanoma site. The overall disease control rate was 50% but varied significantly by KIT mutation status (77% mutated v 18% amplified). Four patients harbored pretreatment NRAS mutations, and one patient acquired increased KIT amplification after treatment. Conclusion Melanomas that arise on mucosal, acral, or CSD skin should be assessed for KIT mutations. Imatinib can be effective when tumors harbor KIT mutations, but not if KIT is amplified only. NRAS mutations and KIT copy number gain may be mechanisms of therapeutic resistance to imatinib.

The potential for guarantee-time bias (GTB), also known as immortal time bias, exists whenever an analysis that is timed from enrollment or random assignment, such as disease-free or overall survival, is compared across groups defined by a classifying event occurring sometime during follow-up. The types of events associated with GTB are varied and may include the occurrence of objective disease response, onset of toxicity, or seroconversion. However, comparative analyses using these types of events as predictors are different from analyses using baseline characteristics that are specified completely before the occurrence of any outcome event. Recognizing the potential for GTB is not always straightforward, and it can be challenging to know when GTB is influencing the results of an analysis. This article defines GTB, provides examples of GTB from several published articles, and discusses three analytic techniques that can be used to remove the bias: conditional landmark analysis, extended Cox model, and inverse probability weighting. The strengths and limitations of each technique are presented. As an example, we explore the effect of bisphosphonate use on disease-free survival (DFS) using data from the BIG (Breast International Group) 1-98 randomized clinical trial. An analysis using a naive approach showed substantial benefit for patients who received bisphosphonate therapy. In contrast, analyses using the three methods known to remove GTB showed no statistical evidence of a reduction in risk of a DFS event with bisphosphonate therapy.

2007Background: Approximately 5-8% of patients with cancer develop leptomeningeal carcinomatosis (LMD). Median survival of patients with LMD is approximately 4-6 weeks and there are no effective tr...

e14041Background: The genetic alterations that define PCNSL are beginning to be identified, and the genomic evolution that conveys susceptibility or resistance to treatment in recurrent disease is ...

2020Background: The impact of individualized neoepitope vaccination targeting neoantigens arising from tumor-specific mutations for GBM, a low mutation burden tumor with an immunologically cold tum...

269 Background: Cholangiocarcinoma (CC) is a fatal malignancy with an unmet treatment need. With the approval of immunotherapy for solid tumors with mismatch repair (MMR) deficiency, there is a renewed interest in MMR testing. Little is known about the incidence of MMR deficiency in CC or its correlation to survival, immune cell infiltration, PD-L1 and other proteins expressed in CC such as mesothelin. Methods: CC tumors were identified from patients treated at the University of Virginia from 2000-2014. Tissue microarrays (TMAs) were constructed of 3-4 cores from each tumor and were stained by immunohistochemistry for MMR genes (MLH1, PMS2, MSH2, MSH6), mesothelin, PD-L1 and immune cells. TMAs were scanned using the Leica SCN400 and analyzed using the Digital Image Hub software. Stain intensity thresholds for defining positive cells were determined by two users and recorded as an average of all cores from each tumor. Mesothelin and PD-L1 expression were measured as a percentage of positive tumor cells. Correlation with overall survival was assessed using log-rank tests and classification and regression trees, with p values < 0.05 considered significant. Results: Ninety-one tumors were analyzed: 24 intrahepatic, 33 hilar, and 34 distal. MMR deficiency was found in 20 tumors (22%). None of the MMR deficient tumors co-expressed PD-L1 (>1%), which was found in 15% of the remaining tumors. T cell infiltration (CD4, CD8 and FoxP3) did not differ between MMR deficient or proficient tumors. Patients with MMR deficiency had a trend towards worse survival compared to those with proficiency (median OS: 19.2 vs. 28.1 months, p = 0.07). MMR deficient tumors showed a lower mesothelin expression compared to MMR proficient tumors, median 8 vs. 129 positive cells per TMA (p = 0.08). Patients with MMR deficiency and low mesothelin expression had a worse outcome compared to patients with MMR proficiency and high mesothelin expression (median OS: 14.5 vs. 30.0 months, p = 0.05). Conclusions: Given the high rate of MMR deficiency, all CC tumors should be tested and may benefit from anti-PD-1 therapy. The poor prognosis of MMR deficient CC may be independent of T-cell infiltration and additional studies are needed to better characterize the genetic and molecular landscape of this subset of tumors.

e14565 Background: Checkpoint inhibition (CPI) has significant and durable effectiveness against a broadening range of cancers and currently is FDA approved for more than 10 different malignancies. With increasing use of CPI, there is a need to understand the range and extent of immune-related adverse effects (irAE). Though certain irAEs, such as colitis and pneumonitis have been well studied, CPI- induced arthritis (CA) has not been widely recognized nor well characterized with data on the subject largely comprised of case reports and small case series. Methods: We retrospectively reviewed 125 patients (pts) with advanced melanoma treated with CPI to define the incidence and clinical features of CA. We identified 20 pts who developed CA during the course of CPI treatment. Cases were included only if active rheumatic signs or symptoms developed after receiving therapy. Demographic data (gender, age), type of CPI, number of CPI cycles, treatment of CA and other irAE manifestations were extracted from the medical chart. Results: 16% of patients treated with CPI developed CA. Among the 20 pts, the average age at CA onset was 70.3 (43, 84) with 75% of pts male. 80% had received anti-PD-1 monotherapy, 15% combination anti-PD-1 + anti-CTLA-4 and 5% monotherapy with CTLA-4 inhibition. The average number of infusions prior to symptom onset was 13.5 (1-48) after a median time point of 6.6mths on therapy. Basic rheumatologic testing (RF, CCP, ANA) was only positive in 10% of cases. 20% of pts were treated with non-steroidal anti-inflammatory drugs alone, whereas 40% required corticosteroids (systemic or injection). The remaining 40% necessitated treatment with disease modifying anti-rheumatic drugs. Conclusions: CA is a debilitating irAE with a higher incidence (16%) than most other irAEs and a negative impact on the quality of life of pts affected. Based on our retrospective review, this irAE tends to delayed in onset and often requires systemic immune suppressants for effective treatment. We posit that this is an under recognized and undertreated irAE. Further work is needed to better define the clinical factors that may affect or predict its development.

9515 Background: CheckMate 064 (open label, phase 2) randomized advanced melanoma pts to NIVO 3 mg/kg Q2W for 6 doses then IPI 3 mg/kg Q3W for 4 doses (n = 70; cohort A), or the reverse (IPI then NIVO; n = 70; cohort B). More cohort A than cohort B pts had complete or partial response (54% and 31%, respectively). We sought to determine whether common or distinct baseline biomarkers were associated with response for each cohort. Methods: Protein expression of biomarkers identifying a broad array of cell lineages and immunoregulatory factors was determined by IHC and evaluated by 2 pathologists, with concordance. Global transcript expression was determined by RNAseq from frozen tissue. An elastic net predictor was trained on cohort A (α = 0.1, λ = 0.05) using leave-one-out cross-validation. RECIST v1.1 was used to assess tumor response. Results: Patients with tumors showing a pro-inflammatory transcriptional signature at baseline had superior overall response in cohort A, but not cohort B. Consistent with this observation, a parsimonious classifier of 10 immune/IFNγ-related genes predicted BOR (AUC 0.87) and OS (CPH log-likelihood P= 0.0003) for patients in cohort A, but not cohort B. In contrast, low baseline expression of MHC class I protein by malignant cells ( < 50% positive) was associated with disease progression (MHCI < 50% vs ≥50%: 12/13 vs 16/28 progressed) and inferior OS for patients in cohort B (MHCI < 50%: 6.42 mo [2.20, 9.66]; MHCI ≥50%: mOS 18.02 mo [7.98, not reached]: P= 0.0021, log-rank test), but not cohort A. Conclusions: Different pre-treatment biological characteristics of melanoma are associated with clinical response to NIVO followed by a planned switch to IPI (cohort A) and clinical response to IPI followed by NIVO (cohort B). Improved outcome with initial NIVO is strongly associated with a pro-inflammatory signature enriched for IFNγ targets. Inferior outcome with initial IPI is associated with reduction/loss of MHC I antigen presentation machinery. These data imply that NIVO broadly activates innate and adaptive immunity, whereas IPI is reliant upon CD8/MHC class I mediated adaptive immunity to effect clinical response. Clinical trial information: NCT01783938.

e21544 Background: Immune checkpoint inhibitors (ICIs) have a promising activity in variety of malignancies however little is known regarding their efficacy and toxicity profile in older adults (≥65 yo). Accordingly, we performed a meta-analysis of published trials using anti-PD-1 or PD-L1 to define the activity of these agents in this subset of patients. Methods: We conducted a search using PubMed and Web of Science databases for randomized controlled trials (RCTs) of PD-1 and PD-L1 inhibitors including nivolumab (N), pembrolizumab (P), and atezolizumab (A). We then identified the studies that reported a subgroup analysis of overall survival (OS) based on an age cutoff. Results: We identified a total of 17 RCTs that compared N, P and A to chemotherapy or targeted therapy. Out of those trials only 9 reported hazard ratio (HR) based on age and were included in this meta-analysis. These trials treated a total of 5033 patients including 2315 patient ≥65 yo (46%). The overall estimated, random-effects hazard ratio (HR) for death was 0.69 with 95% CI of 0.63 -0.74 (p < 0.0001) in the general population favoring ICIs. Based on the selected trials, there was evidence of a 31% reduction in the hazard of death with PD1/PDL1 inhibitors compared with chemo/targeted therapy. This OS benefit persisted when the analysis was done based on disease type (NSCLC or others) or the drug target (PD-1 or PD-L1). Analysis based on age cutoff of 65 yo showed that the efficacy of PD-1 and PD-L1 inhibitors did not vary for older or younger patients. The overall estimated random-effects HR was 0.64 with 95% CI of 0.54 -0.76 in patients ≥65 yo vs. 0.68 with 95% CI of 0.61-0.75 in patients < 65 yo. Data regarding toxicity based on age was available only for 3 trials. The rate of grade 3-4 toxicities was comparable between the 2 groups (45-50% in patients ≥ 65 compared to 37.1-50% in patients < 65). Conclusions: PD-1 and PD-L1 inhibitors efficacy using N, P or A is comparable in older and younger adults when using an age cutoff of 65 years. Future ICIs trials should report efficacy and toxicity based on age.

564Background: Women who exercise after breast cancer diagnosis have a lower risk of recurrence and mortality, but the biological mechanisms through which exercise impacts breast cancer are unclear. The Pre-Operative Health and Body (PreHAB) Study was a randomized window of opportunity trial designed to test the impact of exercise on tissue and serum biomarkers in women with early breast cancer. Methods: Inactive women with newly diagnosed Stage I-III breast cancer were enrolled through Dana-Farber Cancer Institute and Yale University prior to surgery. Participants were randomized 1:1 to an exercise intervention or Mind-Body control intervention and participated in the interventions between enrollment and the time of surgery. Tumor tissue and serum were collected at enrollment and surgery. Results: A total of 49 women were randomized (27 exercise and 22 control). At baseline, mean age was 52.6, BMI was 30.2kg/m2 and exercise was 49 min/wk. Mean time between enrollment and surgery was 4.2 weeks. Exercise p...

9521Background: PD-1 inhibitors are promising anti-cancer agents and are associated with unique immune-related response patterns. We characterized patterns of tumor response dynamics on CT scans du...

9577Background: Despite its rising incidence and aggressive behavior, MCC remains an orphan disease with no active agents. Immunosuppression, UV exposure and MCPyV are known risk factors. A better ...

2071Background: Approximately 50% of pts with metastatic melanoma (MM) develop BM. Among MM pts, pembro, a humanized antibody against PD-1, is associated with objective response rates of 32-35% in ...

167 Background: Evidence increasingly links obesity to increased risk of cancer recurrence and mortality in breast and other cancers, but few studies have evaluated weight loss interventions in cancer patients. We evaluated the impact of a group-based weight loss intervention implemented through an oncology clinic on weight and other outcomes in a mixed population of cancer survivors. Methods: Overweight and obese cancer survivors were randomized 1:1 to immediate or delayed participation in a 15-week group-based weight loss program focused on calorie restriction and increased physical activity. Weight, body composition, physical activity, fitness and quality of life were assessed at baseline and 15 weeks. Changes in measurements between baseline and 15 weeks were compared using Wilcoxon rank sum tests. The primary outcome was change in weight between baseline and 15 weeks between groups. Results: 60 participants were randomized; 30 to intervention and 30 to control. Median age was 52, average BMI was 31.8 kg/m2, 97% of participants were women, and 80% had breast cancer. Intervention participants lost 5.3% of baseline weight at 15 weeks vs 0.2% weight gain in controls (P < 0.001) (Table). Improvements in fitness (6-minute walk test) and physical functioning (EORTC QLQ C30) were also observed in the intervention group vs. controls. Conclusions: We found thatparticipation in a 15-week group-based intervention resulted in weight loss and improvements in fitness and physical functioning in overweight and obese cancer survivors. More work is needed to evaluate the feasibility and sustainability of weight loss programs implemented through oncology practices. Clinical trial information: NCT01978899. [Table: see text]

e20025 Background: The combination of ipilimumab (IPI) 10 mg/kg with GM-CSF (GM) demonstrated an improvement in overall survival and safety profile over IPI alone in a randomized phase II trial. No...

e20074 Background: There is growing evidence of potential synergy of BRAF-targeted therapy and immunotherapy. This phase II trial was conducted to determine the progression-free (PFS) and overall s...

TPS1138 Background: Patient-reported outcomes (PROs) independently and directly describe patient (pt) treatment experiences, are complementary to clinician reports, and are increasingly included in...

9084 Background: Physical activity (PA) interventions improve fitness, functional capacity, and quality of life in patients with early-stage breast cancer. There are almost no data regarding the feasibility or potential benefits of PA in women with metastatic breast cancer (MBC). Our study evaluated the impact of a moderate-intensity PA intervention upon functional measures in patients with MBC. Methods: Participants were randomized 1:1 to a 16-week PA intervention or usual care control group. Eligibility included diagnosis of MBC, ECOG 0-1, and no active brain metastases. The intervention consisted of 4 weekly meetings with an exercise physiologist, followed by monthly in-person and weekly telephone-based meetings. The PA goal was 150 minutes per week. Baseline and 16-week evaluation included: modified Bruce treadmill test, 7-Day Physical Activity Recall Interview, and EORTC QLQ C-30 questionnaire. Results: 101 women enrolled between 9/06 and 3/11. Median age was 49 years, median time since diagnosis of MBC was 1.1 years, 48% of participants were on hormonal therapy and 42% on chemotherapy/biologic therapy. 69% completed all study measures (an additional 9% all but the final treadmill test). Attrition was higher in the intervention arm (30% vs. 16%, p=0.15). Outcome measures are shown in table below. Both groups experienced improvements in treadmill times. Intervention participants also reported improvements in physical functioning and increased minutes of weekly PA, with a trend toward significant differences between groups. Conclusions: Women with MBC participating in a PA intervention experienced consistent, but non-significant, trends toward improvements in functional measures. Our sample size was small, limiting power to evaluate the potential benefits of PA in women with MBC. Additional work is warranted, particularly since PA interventions appear feasible in this patient population. [Table: see text]

8569 Background: Treatment of patients (pts) with BRAF mutant malignant melanoma (MM) with vemurafenib (vem), a BRAF inhibitor (BRAFi), is associated with a high response rate (RR) and improvement in progression free survival (PFS) and overall survival (OS) compared to standard chemotherapy. Responses to vem are typically not durable and most pts require additional therapy. However, there is little data to guide sequencing of vem with other standard therapies such as ipilimumab (ipi) or interleukin 2 (IL-2). Methods: 43 pts treated with vem as part of clinical trials from 2009-2012 were retrospectively identified. RR of vem was calculated for all pts as well as for pts treated with immunotherapy (IT) prior to vem. The OS from first systemic and vem treatment, duration of vem therapy, OS from when pts came off vem, and PFS and OS for post-vem ipi was determined using Kaplan-Meier estimates. Results: Of the 43 pts, 11 remain on vem and 32 are off vem (19 deceased) with a median follow up of 19 mo. Pre-vem LDH was elevated in 46% (17/37 evaluable pts). The RR to vem was 52% (22/42 evaluable pts), the median duration of therapy 5.6 months (mo), and the median OS 19.3 mo, similar to results in trials. The median OS from initiation of any treatment was 27 mo. 16 pts received IT (11 IL-2, 4 ipi, 1 IL-2 then ipi) prior to vem with a median OS of 31.2 mo; the RR of vem following IT was 75% (12/16). At time of vem discontinuation 50% of pts (12/24 evaluable) had an elevated LDH, and the median OS of the 32 pts who stopped vem was 4 mo from last vem dose. 10 of these pts then received single-agent ipi; six of whom had an elevated LDH. Five pts died within 3 mo of last vem dose, and all 10 pts had disease progression (PD) at 6 months with a median PFS of 0.7 mo and OS of 2.2 mo. The 3 pts who lived beyond 1 year following vem and then ipi were subsequently treated with a BRAFi following PD on ipi. Conclusions: Response to vem following IT appears similar to that seen in previously untreated pts. Median PFS and OS for pts receiving ipi after discontinuation of vem are poor, possibly due to rapid PD at the time of vem discontinuation. Prolonged OS is primarily seen with resumption of BRAF inhibition. Our data suggests that in appropriately selected pts, IT should be considered prior to BRAFi.

522 Background: Conflicting data exist regarding the association of side effects related to estrogen manipulation with prognosis. We evaluated the outcome of patients (pts) according to the early i...

8511 Background: CTLA-4 blockade with ipilimumab demonstrated a survival advantage in advanced melanoma. Ipilimumab induces an immune mediated tumor vasculopathy. In addition to angiogenic effects, VEGF is a potent immune suppressor of antigen presenting cells. Therefore, we performed the first combination study to investigate potential synergies of ipilimumab and bevacizumab. Methods: Patients (pts) were eligible with stage III unresectable or stage IV disease, no bowel or CNS metastases, ECOG PS 0 or 1, > 4 wks from prior therapy, adequate end organ function, no autoimmune or bleeding problems. Pts received ipilimumab every 3 weeks IV x 4 then q12 weeks, and bevacizumab q 3 weeks. Pts could continue if good PS with up to 40% increase in the sum of the longest diameter and no more than 2 new target lesions. Cohort 1 comprised 10 mg/kg ipilimumab plus 7.5 mg/kg bevacizumab. If ≥3/5 pts did not experience DLT then cohort 2 could be enrolled at 10 mg/kg ipilimumab plus 15 mg/kg bevacizumab. An additional 12...

545 Background: Osteoporosis and fractures are long term complications of aromatase inhibitor use in the adjuvant therapy of breast cancer. Early detection of patients (pts) at risk of treatment-induced osteoporosis may allow preventive therapy and selection of the most appropriate endocrine therapy. We developed a statistical model describing the evolution of BMD in pts treated with T, L and sequences of T and L. Methods: Available dual-energy x-ray absorptiometry exams (DXA) of pts treated in trial BIG 1–98 were retrospectively collected from Swiss centers. Treatment arms: A) T for 5 years; B) L for 5 years; C) 2 years of T followed by 3 years of L; and D) 2 years of L followed by 3 years of T. Pts without DXA were used as a control for detecting selection bias. A repeated measures model using the first-order autoregressive covariance structure to allow for different times between DXA was used to model BMD (g/cm2) since trial randomisation. Prospectively defined covariates were considered as fixed effects in a multivariable model using an intention to treat analysis. Covariates at trial randomization were: age, height, weight, race, known osteoporosis, tobacco use, prior bone fracture, prior hormone replacement therapy (HRT), bisphosphonate use and previous neo-/adjuvant chemotherapy (ChT). Results: A total of 247 out of 546 pts had between 1 and 5 DXA; a total of 576 DXA were collected. Arm A contained 67 pts, B 63 pts, C 55 pts and D 62 pts. Median follow-up was 5.8 years. Factors correlated with BMD in the multivariate analysis were weight (0.003/kg, p < 0.0001), height (0.003/cm, p = 0.0083), osteoporosis (-0.130, p < 0.0001), tobacco (current / previously vs. never: -0.057, p = 0.0011 / -0.042, p = 0.0798), previous HRT (0.030, p = 0.0244), ChT (0.032, p = 0.0174), time since endocrine therapy start (-0.009/year, p = 0.0164) and treatment arm (B / C / D vs. A: -0.068, p = 0.0002 / -0.091, p < 0.0001 / -0.064, p = 0.003). Conclusions: Our statistical model describes the BMD evolution for pts treated with L and/or T. All treatment regimens affect BMD. Contrary to expectation, the switch schedule T followed by L does not seem to result in better bone protection compared to L monotherapy. [Table: see text]