Niklas Gebauer

Introduction: Pemetrexed, an antifolate, is a standard therapy for lung adenocarcinoma (LUAD). Since folic acid is vital for nucleic acid metabolism, disrupting its pathway affects both tumor cells and regenerating healthy cells. The PemToxTrial examined the impact of pemetrexed on severe hematotoxicity and mortality. Methods: This retrospective analysis included 136 LUAD patients (UICC8 stage IVA/B) receiving palliative treatment. Ninety-two patients received pemetrexed-based therapy, while 44 received pemetrexed-free treatment (68 and 38 with immune-checkpoint inhibitors). Bone metastasis (BM) was present in 47 patients in the pemetrexed group and 22 in the comparison group. The primary endpoint was grade III–V hematotoxicity; secondary endpoints included mortality, progression-free survival (PFS), and overall survival (OS). Hematologic outcomes following a subsequent change in clinical practice were prospectively assessed in a post-implementation cohort of 15 patients with bone-metastatic LUAD receiving pemetrexed-free regimens. Results: Median age was 66 years, 59.6% were male and median follow-up was 13.5 months. Grade III-V hematotoxicity was most frequent (81%) in patients with BM receiving pemetrexed (n = 47), significantly higher than in other subgroups (p < 0.0001). Mortality was also increased in this group (66% vs. 29%, p < 0.0001), with hematotoxicity directly contributing to death in 42% versus 11.5% of cases. PFS was significantly inferior (5.2 vs. 8.6 months). Multivariable analysis identified BM as the only independent predictor of severe hematotoxicity and mortality. Only a mutational burden of ≥4 potential driver alterations was associated with inferior PFS. Bone-derived total lesion glycolysis (TLG) on PET-CT strongly correlated with high-grade hematotoxicity (p < 0.0001). Conclusions: The PemToxTrial identified a strong association between pemetrexed exposure, bone metastases, and severe hematotoxicity, supporting further prospective evaluation of bone metastasis status as a potential clinical risk factor for treatment-associated hematotoxicity.

For patients with large B-cell lymphoma progressing after multiple lines of therapy few treatment options remain, including the new CD19-directed antibody-drug conjugate loncastuximab tesirine (lonca). To further elucidate its efficacy and safety, we conducted a real-world analysis including data from 31 German centers. Ninety-three heavily pre-treated patients received lonca in third and later line of therapy. 76% had received treatment with chimeric antigen receptor T-cells (n=55), bispecific antibodies (BsAbs) (n=56) or both (n=40). The overall response rate was 27% and complete response rate was 10%. Median progression-free (mPFS) and overall survival (mOS) were 2.2 and 4.4 months, respectively. Median PFS of responders was 11.4 months. Toxicity was acceptable, with grade ≥ 3 infections (13%) being the most common adverse event. Patients receiving subsequent allogeneic stem cell transplantation survived significantly better than the remaining non-consolidated patients (mOS: 15.2 vs. 3.8 months, p=0.048). In multivariable analysis, no response to the last therapy prior to lonca (OS - HR: 2.5, p=0.014) and secondary IPI (PFS/OS - HR: ≥2.0, p≤0.024) were significantly associated with poor outcomes. Lonca demonstrated a favorable safety profile with moderate efficacy in heavily pretreated patients. While enabling timely access to subsequent therapy, efficacy in earlier lines with combination partners may further increase response rates and durability of response.

EBV-positive diffuse large B-cell lymphoma (DLBCL) is an aggressive lymphoma enriched in elderly patients, in which PD-L1 genomic gains cooperate with viral signalling to promote immune escape. However, the spatial mechanisms by which PD-L1 reshapes the tumour microenvironment remain unresolved. Here, we integrate genomic profiling with spatial transcriptomics, spatial proteomics, and single-cell RNA sequencing to resolve the tumour-immune architecture of EBV-positive DLBCL. PD-L1 gain tumours exhibit a striking decoupling of immune proximity and immune engagement: T cells accumulate in close proximity to tumour cells yet direct tumour-immune contact is largely absent. Tumour boundaries are enriched for cancer-associated fibroblasts and display transcriptional signatures of metabolic immune suppression, while tumour-proximal T cells exhibit pronounced exhaustion. Together, these findings reveal a PD-L1 driven spatial immune evasion architecture that permits immune infiltration while enforcing multilayered suppression through fibroblast-mediated exclusion, metabolic constraint, and T-cell dysfunction, ultimately preventing effective anti-tumour immunity in EBV-positive DLBCL.

Cancer treatment has shifted toward personalized therapy based on molecular profiling, particularly in advanced disease. Existing circulating tumor DNA panels are often broad, generating many non-actionable variants and incurring costs that limit routine use in molecular tumor boards. We developed and validated a manufacturer-independent, 109-gene liquid biopsy-centered pan-cancer open next generation sequencing panel (LION panel), combined with an in-house bioinformatic pipeline to support clinical decision-making. A total of 87 samples were analyzed, including 17 reference samples, 21 healthy blood donor controls, and 49 patient samples including nine tumor entities. The LION panel achieved 92% sensitivity and 99% specificity in reference samples, with high concordance to digital droplet PCR (r = 0.99). It detected variant allele frequencies as low as 0.05% (tumor-informed) and 0.5% (tumor-uninformed). Clinical concordance reached 82% with blood-based digital droplet PCR and 75% with whole exome tissue sequencing. In representative cases, variant dynamics correlated with disease progression and revealed additional targetable variants. Overall, the LION panel supports clinical decision-making by enabling identification of targetable variants, disease monitoring, and detection of treatment resistance, particularly when tumor tissue is unavailable.

Background: High-grade soft tissue sarcomas (STSs) are heterogeneous tumours lacking robust prognostic or predictive biomarkers. Regnase-1, an immune RNase, enhances antitumour immunity by limiting immunosuppressive tumour microenvironment (TME) components (e.g., myeloid-derived suppressor cells (MDSCs)), but remains unexplored in STS. As CD68⁺ tumour-associated macrophages (TAMs) drive TME suppression and poor prognosis in non-translocation-driven STS, we evaluated Regnase-1 and CD68⁺ TAMs to assess Regnase-1 as an indicator of an immunologically activated TME. Methods: Immunohistochemistry scoring of Regnase-1 and CD68⁺ TAMs was performed in 91 patients. Overall survival (OS) was assessed by Kaplan–Meier and Cox regression, and findings were validated in an independent “The Cancer Genome Atlas” Sarcoma (TCGA-SARC) cohort (n = 212). Results: In UPS, Regnase-1-high predicted longer OS (17.0 months vs. not reached; p = 0.0247) and lower mortality (univariate hazard ratio (HR) = 0.3; p = 0.0343; multivariate HR = 0.4; p = 0.0413), but not after radiotherapy. CD68⁺ TAM-high predicted shorter OS (13.0 months vs. not reached; p = 0.0274) and higher mortality (HR = 2.0, 95% CI 1.1–3.7; p = 0.0325). Both Regnase-1 effects were reproduced in TCGA-SARC. Regnase-1-high tumours showed inflammatory/interferon enrichment, reduced TGF-β signalling, and SERPINE1 upregulation. Conclusions: Regnase-1 marked a pro-inflammatory TME and favourable outcome in UPS, but this effect may reverse upon radiotherapy.

Introduction: Brain metastases represent a major clinical challenge due to a distinct immunosuppressive microenvironment and limited, heterogeneous efficacy of PD-1/PD-L1 immune checkpoint inhibition. The adenosine pathway, mediated by the ectonucleotidases CD39 and CD73, has emerged as an alternative immune escape mechanism, yet its relevance in brain metastases across tumor entities remains insufficiently characterized. Methods: We conducted targeted panel sequencing of brain metastases from multiple primary tumor entities and evaluated compartment-resolved expression of CD39, CD73, and PD-L1 by immunohistochemistry, distinguishing tumor cell and immune cell expression. Tumor mutational burden (TMB), recurrent gene alterations, and gene fusions were analyzed and integrated with immune marker profiles to define immunogenomic subtypes. Results: Brain metastases displayed a heterogeneous mutational landscape with recurrent alterations including TP53, KRAS, PIK3CA, and APC. CD39 and CD73 expression was frequent and highly variable, occurring on both tumor cells and tumor-infiltrating immune cells, and only partially overlapping with PD-L1 expression. A substantial subset of cases exhibited an adenosine-high phenotype despite low or absent PD-L1. Marker-associated enrichment analyses identified distinct genetic correlates, including enrichment of KRAS alterations in tumors with CD39/CD73 positivity on malignant cells, and APC/PIK3CA-associated patterns linked to immune compartment marker expression. TMB did not significantly differ across major tumor entity groups. Gene fusions were detected in a subset of tumors but were largely independent of immune phenotypes. Conclusions: Adenosine pathway activation is a frequent, genetically associated immune escape feature of brain metastases that complements PD-L1-based stratification. Integrating CD39/CD73 with PD-L1 enables actionable immunogenomic subtyping and supports rational immunotherapy strategies targeting adenosine-mediated immunosuppression.

Background: High-grade soft tissue sarcoma (STS) are heterogeneous tumours lacking robust prognostic or predictive biomarkers. Regnase-1, an immune RNase, enhances antitumour immunity by limiting immunosuppressive tumour microenvironment (TME) components (e.g., myeloid-derived suppressor cells (MDSCs)), but remains unexplored in STS. As CD68⁺ tumour-associated macrophages (TAMs) drive TME suppression and poor prognosis in non-translocation-driven STS, we evaluated Regnase-1 and CD68⁺ TAMs to assess Regnase-1 as an indicator of an immunologically activated TME. Methods: Immunohistochemistry scoring of Regnase-1 and CD68⁺ TAMs was performed in 91 patients; OS was assessed by Kaplan–Meier and Cox regression; findings were validated in an independent TCGA-SARC cohort (n = 212). Results: In UPS, Regnase-1-high predicted longer OS (17.0 months vs. not reached; p = 0.0247) and lower mortality (univariate HR = 0.3; p = 0.0343; multivariate HR = 0.4; p = 0.0413), but not after radiotherapy. CD68⁺ TAM-high predicted shorter OS (13.0 months vs. not reached; p = 0.0274) and higher mortality (HR = 2.0, 95% CI 1.1–3.7; p = 0.0325). Both Regnase-1 effects were reproduced in TCGA-SARC. ZC3H12A-high tumours showed inflammatory/interferon enrichment, reduced TGF-β signalling, and SERPINE1 upregulation. Conclusions: Regnase-1 marked a pro-inflammatory TME and favourable outcome in UPS, but this effect may reverse upon radiotherapy.

Patients with large B-cell lymphoma (LBCL) who experience relapsed disease after CD19-directed chimeric antigen receptor (CAR) T-cell (CAR-T) therapy have a poor prognosis. Bispecific antibodies (BsAbs) induce complete remissions in ∼35% of these cases. Hypothesizing overlapping LBCL-intrinsic resistance mechanisms as well as common poor prognosis predictors to CAR-T and BsAb therapy, we conducted a multicenter retrospective analysis including 92 patients with relapsed/refractory (R/R) LBCL treated with BsAbs after CAR-T failure. Overall response rate (ORR) was 43%, with a progression-free survival (PFS) of 2.8 months. Patients receiving BsAbs during early relapse (≤3 months) achieved a significantly worse outcome (ORR, 29%; PFS, 2.2 months) compared with patients with an intermediate (4-6 months; ORR, 54%; PFS, 3.7 months) or a late relapse (>6 months; ORR, 60%; PFS, 10.5 months). The benefit of later relapse was particularly notable in patients receiving BsAbs as first salvage therapy compared with those receiving a BsAb in subsequent lines (PFS not reached vs 2.7 months; overall survival not reached vs 9.1 months, respectively). In addition to early R/R state before BsAbs, elevated lactate dehydrogenase and higher International Prognostic Index score were significant predictors of poor outcomes to BsAb in multivariate Cox regression analyses. The finding that patients with early relapse after CAR-T respond particularly poorly to BsAb highlights the necessity for alternative treatment options in this high-risk patient cohort.