BackgroundCurrent guidelines recommend that patients with HER2-low metastatic breast cancer (MBC) receive sequentially two antibody-drug conjugates (ADCs): Sacituzumab Govitecan (SG) and Trastuzumab Deruxtecan (T-DXd), despite a similar payload. However, the effectiveness of one after another is unknown.MethodsADC-Low is a multicentre, retrospective study evaluating the efficacy of SG and T-DXd, one after another, with or without intermediary lines of chemotherapy, in patients with HER2-low MBC.ResultsOne hundred and seventy-nine patients were included: the majority with HR-negative tumours received SG first (ADC1) (n = 100/108) while most with HR-positive tumours received T-DXd first (n = 56/71). Median progression-free survival 2 was short: 2.7 months (95% CI: 2.4-3.3) in the whole population, respectively, 3.1 (95% CI: 2.6-3.6) and 2.2 months (95% CI: 1.9-2.7) for patients receiving T-DXd or SG second (ADC2). Intermediary lines of chemotherapy between ADC1 and ADC2 had no impact. Primary resistance to ADC2 occurred in 54.4% of patients. Certain patients showed initial response to ADC2.ConclusionsClinical benefit of sequentially administered SG and T-DXd is limited for most patients. Nevertheless, a subset of patients might benefit-on the short term-from a second ADC. Additional studies are needed to identify patients who could benefit from two ADCs with similar payloads.

Objectives: To assess the use of molecular genotyping to accurately diagnose and treat human chorionic gonadotropin (hCG)-producing tumors and to evaluate the discriminating capacity of molecular testing on prognosis and overall survival.Methods: We conducted a retrospective descriptive study of patients registered with the French Reference Center for Trophoblastic Disease between 1999 and 2021. We included all patients with hCG-producing tumors for whom results of molecular genotyping were available.Results: Fifty-five patients with molecular genotyping were included: 81.2 % (n = 45) had tumors of gestational origin, 12.7 % (n = 7) of non-gestational origin and 5.5 % (n = 3) of undetermined origin. The results of molecular genotyping influenced the treatment decisions for 17 % of patients in this cohort. Overall survival was 93.3 % for patients with gestational tumors (after a median follow-up of 74 months) compared to 71.4 % for patients with non-gestational tumors (after a median follow-up of 23 months).Conclusion: In atypical presentations of hCG-producing tumors, molecular genotyping is a valuable tool to guide diagnosis and tailor treatment recommendations.

Abiraterone acetate (AA) combined with prednisone (P) is a major treatment for metastatic castration-resistant prostate cancer (mCRPC). In case of resistance to AA+P other drugs such as docetaxel and enzalutamide can be used. The combination of AA and dexamethasone (D) has been described as another possible option, immediately (switch) or in rechallenge after a different treatment line. 77 patients with mCRPC were included in this retrospective study conducted in two centers. All patients received AA + P. After disease progression on AA+P, patients were treated either first with another drug and then rechallenge of AA combined with D or with an immediate switch from P to D. Progression was defined by PCWG3 criteria. The efficacy of AA+D was estimated by the proportion of patients with a PSA decline > 50% from baseline value (PSA50), a PSA decline > 30% from baseline value (PSA30) and progression-free survival (PFS). Median duration of treatment with AA + D was 533 days. 22/77 (28.6%) patients had a PSA50 (90% CI: 20.2-38.2) and 28/77 (36.3%) PSA30 (90% CI: 27-46). The biological response rate was no different whether the patients were treated in a switch or rechallenge context. Patients with prior response on AA + P were more likely to achieve PSA50 on AA + D, but the results were not significant: 17/77 (34.6%) versus 4/77 (16%, p = 0.09). Age was found as a predictive factor for PSA50: OR 0.93 (CI 97.5%: 0.87-0.99; p < 0.05). Median PFS on AA+D was 4.2 months. PFS was similar after either switch or rechallenge. Treatment with AA+D was well tolerated, with no significant mineralocorticoid syndrome and no grade ≥3 adverse events. Directly switch from P to D after progression or AA+D rechallenge may be an effective treatment in men without clinical deterioration. Data on predictive factors of response will be further presented.

Chemo-induced thrombocytopenia is a limiting adverse event in glioblastoma patients receiving temozolomide (TMZ). Constitutional single nucleotide polymorphisms (SNP) of genes involved in the pharmacodynamics and pharmacokinetics of TMZ were described to be associated with increased risk of TMZ-induced myelotoxicity in retrospective cohorts. We prospectively investigate the risk of thrombocytopenia (<100G/L) regarding MGMT rs2308327 and ABCB1 rs1045642. A total of 100 newly-diagnosed glioblastoma patients, receiving standard treatment (concomitant TMZ and radiotherapy [RT] followed by TMZ maintenance phase), were included in the study. Loss of function allelic variants rs2308327 of the MGMT gene; and rs1045642 of the ABCB1 gene were characterized using Taqman® PCR. Using univariate and multivariate logistic regression we investigated the association between SNPs and TMZ-induced thrombocytopenia occurrence along first-line treatment schedule and overall survival. This study is ancillary to the ongoing prospective phase 2 GLIOPLAK trial (NCT02617745). Overall, 23/100 patients experienced TMZ-induced thrombocytopenia: 11 patients during the RT-TMZ phase and 12 other patients during the TMZ maintenance phase. Among them, 7 patients experienced severe thrombocytopenia <50 G/l. Observed allele frequencies were in accordance with those reported for Caucasians (11% for MGMT rs2308327 and 47.5% for ABCB1 rs1045642). Harboring rs2308327 or rs1045642 was not associated with TMZ-induced thrombocytopenia: odds ratio (OR) 1.1 [CI 95% 0.3-3.2, p=0.56] and OR 1.01 [0.49-2.09, p=0.55] respectively. Severe thrombocytopenia (<50 G/L, n=7 patients) were not associated with any of the two SNPs. Wild-type versus heterozygous or homozygous SNP did not influence overall survival, log-rank test p=0.19 and p=0.34 respectively. Occurrence of thrombocytopenia <100 G/l induced an underexposure to TMZ in maintenance phase: median 3 cycles vs 5 in the no thrombocytopenia group. rs2308327 and rs1045642 were not associated with TMZ-related thrombocytopenia in a prospective population of glioblastoma.