Development Of Brain Metastases Research Articles

RARRES2 regulates lipid metabolic reprogramming to mediate the development of brain metastasis in triple negative breast cancer

BackgroundTriple negative breast cancer (TNBC), the most aggressive subtype of breast cancer, is characterized by a high incidence of brain metastasis (BrM) and a poor prognosis. As the most lethal form of breast cancer, BrM remains a major clinical challenge due to its rising incidence and lack of effective treatment strategies. Recent evidence suggested a potential role of lipid metabolic reprogramming in breast cancer brain metastasis (BCBrM), but the underlying mechanisms are far from being fully elucidated.MethodsThrough analysis of BCBrM transcriptome data from mice and patients, and immunohistochemical validation on patient tissues, we identified and verified the specific down-regulation of retinoic acid receptor responder 2 (RARRES2), a multifunctional adipokine and chemokine, in BrM of TNBC. We investigated the effect of aberrant RARRES2 expression of BrM in both in vitro and in vivo studies. Key signaling pathway components were evaluated using multi-omics approaches. Lipidomics were performed to elucidate the regulation of lipid metabolic reprogramming of RARRES2.ResultsWe found that down-regulation of RARRES2 is specifically associated with BCBrM, and that RARRES2 deficiency promoted BCBrM through lipid metabolic reprogramming. Mechanistically, reduced expression of RARRES2 in brain metastatic potential TNBC cells resulted in increased levels of glycerophospholipid and decreased levels of triacylglycerols by regulating phosphatase and tensin homologue (PTEN)-mammalian target of rapamycin (mTOR)-sterol regulatory element-binding protein 1 (SREBP1) signaling pathway to facilitate the survival of breast cancer cells in the unique brain microenvironment.ConclusionsOur work uncovers an essential role of RARRES2 in linking lipid metabolic reprogramming and the development of BrM. RARRES2-dependent metabolic functions may serve as potential biomarkers or therapeutic targets for BCBrM.

Advances in the application of neuroinflammatory molecular imaging in brain malignancies.

The prevalence of brain cancer has been increasing in recent decades, posing significant healthcare challenges. The introduction of immunotherapies has brought forth notable diagnostic imaging challenges for brain tumors. The tumor microenvironment undergoes substantial changes in induced immunosuppression and immune responses following the development of primary brain tumor and brain metastasis, affecting the progression and metastasis of brain tumors. Consequently, effective and accurate neuroimaging techniques are necessary for clinical practice and monitoring. However, patients with brain tumors might experience radiation-induced necrosis or other neuroinflammation. Currently, positron emission tomography and various magnetic resonance imaging techniques play a crucial role in diagnosing and evaluating brain tumors. Nevertheless, differentiating between brain tumors and necrotic lesions or inflamed tissues remains a significant challenge in the clinical diagnosis of the advancements in immunotherapeutics and precision oncology have underscored the importance of clinically applicable imaging measures for diagnosing and monitoring neuroinflammation. This review summarizes recent advances in neuroimaging methods aimed at enhancing the specificity of brain tumor diagnosis and evaluating inflamed lesions.

Establishment of a prognostic risk prediction modelfor non-small cell lung cancer patients with brainmetastases: a retrospective study.

Patients with non-small cell lung cancer (NSCLC) who develop brain metastases (BM) have a poor prognosis. This study aimed to construct a clinical prediction model to determine the overall survival (OS) of NSCLC patients with BM. A total of 300 NSCLC patients with BM at the Yunnan Cancer Centre were retrospectively analysed. The prediction model was constructed using the least absolute shrinkage and selection operator-Cox regression. The bootstrap sampling method was employed for internal validation. The performance of our prediction model was compared using recursive partitioning analysis (RPA), graded prognostic assessment (GPA), the update of the graded prognostic assessment for lung cancer using molecular markers (Lung-molGPA), the basic score for BM (BSBM), and tumour-lymph node-metastasis (TNM) staging. The prediction models comprising 15 predictors were constructed. The area under the curve (AUC) values for the 1-year, 3-year, and 5-year time-dependent receiver operating characteristic (curves) were 0.746 (0.678-0.814), 0.819 (0.761-0.877), and 0.865 (0.774-0.957), respectively. The bootstrap-corrected AUC values and Brier scores for the prediction model were 0.811 (0.638-0.950) and 0.123 (0.066-0.188), respectively. The time-dependent C-index indicated that our model exhibited significantly greater discrimination compared with RPA, GPA, Lung-molGPA, BSBM, and TNM staging. Similarly, the decision curve analysis demonstrated that our model displayed the widest range of thresholds and yielded the highest net benefit. Furthermore, the net reclassification improvement and integrated discrimination improvement analyses confirmed the enhanced predictive power of our prediction model. Finally, the risk subgroups identified by our prognostic model exhibited superior differentiation of patients' OS. The clinical prediction model constructed by us shows promise in predicting OS for NSCLC patients with BM. Its predictability is superior compared with RPA, GPA, Lung-molGPA, BSBM, and TNM staging.

EVALUATING HUMAN MICROBIOME SIGNATURES ASSOCIATED WITH DEVELOPMENT, OUTCOME, AND TREATMENT RESPONSE IN GLIOMA AND BRAIN METASTASIS: A SCOPING REVIEW

Abstract The human microbiome is made up of over 500 species of bacteria in the body. Microbiome dysbiosis has been implicated in cancer development and treatment response, including in both primary brain tumors and brain metastases, by acting through the gut-brain axis. We conducted a scoping review of the current evidence surrounding the human microbiome and brain tumors. METHODS: A systematic search of relevant studies and abstracts from 5 electronic databases from 1946-2023 was conducted based on a search strategy developed by an information specialist. Eligible studies included human, in vivo, or in vitro studies that focused on the relationship between microbiome and glioma and brain metastasis development, response to therapies, and outcomes. RESULTS: Of 383 citations, 31 studies met inclusion criteria, including 15 articles and 16 conference abstracts. There were 19 human studies and 18 mouse studies, of which 24 studies focused on glioma and 7 studies focused on brain metastases. They either characterized the microbiome in patients with brain tumors (n=19) or its correlation to systemic therapies for glioma (n=12). Changes in the Firmicutes/Bacteroidetes ratio, which is a marker of dysbiosis in patients with brain tumors was observed in 7 studies; however, the direction of change varied. Microbiome dysbiosis seemed to occur in response to both immunotherapy (5 studies) and temozolomide (3 studies), with variable impacts on treatment response. CONCLUSION: Evidence surrounding the impact of the gut-brain axis on brain tumors remains in its infancy with a large amount of heterogeneity.

Impact of AKT1 on cell invasion and radiosensitivity in a triple negative breast cancer cell line developing brain metastasis.

The PI3K/AKT pathway is activated in 43-70% of breast cancer (BC)-patients and promotes the metastatic potential of BC cells by increasing cell proliferation, invasion and radioresistance. Therefore, AKT1-inhibition in combination with radiotherapy might be an effective treatment option for triple-negative breast cancer (TNBC)-patients with brain metastases. The impact of AKT1-knockout (AKT1_KO) and AKT-inhibition using Ipatasertib on MDA-MB-231 BR cells was assessed using in vitro cell proliferation and migration assays. AKT1-knockout in MDA-MB-231BR cells was performed using CRISPR/Cas9. The effect of AKT1-knockout on radiosensitivity of MDA-MB-231BR cell lines was determined via colony formation assays after cell irradiation. To detect genomic variants in AKT1_KO MDA-MB-231BR cells, whole-genome sequencing (WGS) was performed. Pharmacological inhibition of AKT with the pan-AKT inhibitor Ipatasertib led to a significant reduction of cell viability but did not impact cell migration. Moreover, only MDA-MB-231BR cells were sensitized following Ipatasertib-treatment. Furthermore, specific AKT1-knockout in MDA-MB-231BR showed reduced cell viability in comparison to control cells, with significant effect in one of two analyzed clones. Unexpectedly, AKT1 knockout led to increased cell migration and clonogenic potential in both AKT1_KO clones. RNAseq-analysis revealed the deregulation of CTSO, CYBB, GPR68, CEBPA, ID1, ID4, METTL15, PBX1 and PTGFRN leading to the increased cell migration, higher clonogenic survival and decreased radiosensitivity as a consequence of the AKT1 knockout in MDA-MB-231BR. Collectively, our results demonstrate that Ipatasertib leads to radiosensitization and reduced cell proliferation of MDA-MB-231BR. AKT1-inhibition showed altered gene expression profile leading to modified cell migration, clonogenic survival and radioresistance in MDA-MB-231BR. We conclude, that AKT1-inhibition in combination with radiotherapy contribute to novel treatment strategies for breast cancer brain metastases.

Nivolumab and ipilimumab with concurrent stereotactic radiosurgery for intracranial metastases from non-small cell lung cancer: analysis of the safety cohort for non-randomized, open-label, phase I/II trial

BackgroundUp to 20% of patients with non-small cell lung cancer (NSCLC) develop brain metastasis (BM), for which the current standard of care is radiation therapy with or without surgery. There...

Real-world data of pyrotinib-based therapy for patients with brain metastases of HER2-positive advanced breast cancer: a single-center retrospective analysis and molecular portraits.

Pyrotinib is a novel irreversible pan-HER tyrosine kinase inhibitor (TKI). However, real-world data of pyrotinib-containing therapy in human epidermal growth factor receptor 2 (HER2)-positive metastatic breast cancer (MBC) and developing brain metastases (BMs) are limited, and the genomic profile of this subpopulation is almost undefined. Patients with BM of HER2-positive MBC (n = 35) treated with pyrotinib-containing therapy were enrolled in this analysis. Progression-free survival (PFS), overall survival (OS), objective response rate (ORR), disease control rate (DCR), and toxicity profiles were evaluated. Hazard ratios (HRs) and 95% confidence intervals (CIs) for disease progression were estimated using the Cox proportional hazards models. Targeted next-generation sequencing of 618 cancer-relevant genes was performed on plasma and primary breast tumors from patients with BM and without BM. The median PFS time was 8.00 (95% CI, 5.98-10.017) months, and the median OS time was 23 (95% CI, 10.412-35.588) months. The ORR was 45.7%, and the DCR was 74.3%. In the Cox multivariate analysis, prior exposure to brain radiotherapy (HR = 3.268), received pyrotinib as third- or higher-line treatment (HR = 4.949), subtentorial brain metastasis (HR = 6.222), and both supratentorial and subtentorial brain metastases (HR = 5.863) were independently associated with increased risk of progression. The frequent grade 3-4 adverse event was increased direct bilirubin (14.3%), and two patients suffered from grade 3-4 diarrhea. In the exploratory genomic analysis, altered frequencies of FGFR3, CD276, CDC73, and EPHX1 were higher in the BM group. The consistency of mutated profiles of plasma and primary lesion in the BM group was significantly lower (30.4% vs. 65.5%; p = 0.0038). Pyrotinib-containing therapy shows favorable effectiveness and tolerable safety in patients with BM of HER2-positive MBC, particularly in a population that is brain radiotherapy-naïve, received pyrotinib as first- or second-line treatment, and developed supratentorial brain metastasis. In the exploratory genomic analysis, patients with BM showed distinct genomic features from patients without BM.

Brain Metastasis is Associated with Tumor Size, Nodal Status, and c-erbB-2 Expression in Invasive Breast Carcinoma

Abstract Introduction According to the World Health Organization (WHO) classification, invasive breast carcinoma (IBC) of no special type (IBC-NST) is the second most common primary site of central nervous system metastases, affecting 15% to 30% of patients. Brain metastasis originating from IBC is associated with patient age, tumor size, and axillary lymph node status. Loss of expression of hormone receptors and c-erbB-2 amplification are frequent findings in patients who develop brain metastasis. Radiological studies of the central nervous system are carried out only in patients presenting with neurological signs or symptoms during the clinical follow-up. Objective To evaluate the associations of clinical and pathological findings with brain metastasis in breast cancer. Materials and Methods The sample comprised 73 patients with breast cancer who underwent mastectomy with lymph node resection. The following variables were evaluated: tumor size, histological grade, nodal state, expression of estrogen and progesterone receptors and c-erbB-2, and presence of brain metastasis. Results The histopathological findings associated with brain metastasis in patients with IBC were tumor size (p = 0.03), presence of nodal metastasis (p = 0.045), and c-erbB-2 expression (p = 0.012). Conclusion The assessment of specific pathological findings in breast carcinoma can help identify risk factors and/or clinical parameters associated with the development of brain metastasis.

Real world outcomes of patients treated with first line chemo-immunotherapy (IO) for small cell lung cancer (SCLC): Impact of brain metastases and patterns of subsequent therapy.

e20639 Background: The addition of IO to chemotherapy for first-line treatment of SCLC has improved median overall survival (OS). No OS difference was observed in patients with brain metastases in the clinical trial setting, however outcomes need to be assessed in the non-clinical trial setting. Methods: This retrospective study included 121 patients with SCLC who received ICI-based therapy between 2016-2022 at the Ohio State University. Medical records were extracted via chart review and data was collected in REDCap. Clinical characteristics were summarized using descriptive statistics and compared between patients with and without brain metastasis using Kruskal-Wallis test (continuous outcomes) and Fisher’s exact test (categorical outcomes). Kaplan-Meier survival analysis were used to assess the survival outcomes. All statistical analysis was done in SAS 9.4. Results: Median age at diagnosis was 64.7 years (IQR: 60.1-69.7). Our cohort included 53.8% males and 46.2% females, majority (89.1%) of patients were white, 7.6% were African Americans. Baseline ECOG status was 0 (11.9 %), 1(50%) and > = 2 (38.1%). 97% were current or former smokers. Most had extensive stage disease (96.6%) at diagnosis. Most common sites of metastasis at diagnosis were liver (46.2%), bone (42.9%), brain (35.3%), lymph nodes (34.7%). All patients received first line chemotherapy with atezolizumab (atezo), 40% received atezo with cycle one of chemotherapy, 47% had atezo added with the second cycle, and 13% had atezo added after cycle 2. Median OS for all patients was 9.5 months (95% CI 8.1, 12.9 months); 81% had disease progression at a median follow up of 8.5 months (95% CI 5.2-16.8). 52% patients received second line therapy, 18% received third line and 12% received fourth line and beyond. Subsequent therapies included chemotherapy (29.8%), immunotherapy (4.1%) and clinical trials (14.1%). CNS progression was seen in 47.9%. Median OS for patients with baseline brain metastasis was not significantly different than patients without,11.8 months (95% CI 8-17.6 months) compared to 8.3 months (95% CI 6.2-12.9 months), p:0.26. For patients without baseline brain metastases, median time to developing brain metastasis from starting chemo-IO was 7.3 months [IQR: 5.3-14.1 months]. There was no significant association between the development of brain metastasis and age, sex, or race. At CNS progression 36.4% received SBRT, 18.2% received WBRT. Conclusions: Median overall survival in this cohort was shorter than what has been reported in the phase 3 trials with first line chemo-IO. OS for patients with brain metastases did not differ significantly than for those without, and there was a higher rate of SRS/SBRT versus WBRT. Just over half of patients received subsequent lines of therapy highlighting the need for continued improvement in treatment options for SCLC.

Assessment of electronic health record (EHR) –based machine learning (ML) in predicting risk of brain metastasis among patients with early-stage non–small-cell lung cancer (eNSCLC).

2036 Background: Among patients (pts) with eNSCLC, development ofbrain metastasis (BM) is a poor prognostic sign, but routine surveillance for BM is not recommended post-therapy. EHR-based ML algorithms may identify pts who would benefit from active brain MRI surveillance and/or treatment intensification in the early-stage setting. Methods: ConcertAI Patient360 database, consisting of structured and curated EHR records from pts receiving care in ~900 US oncology clinics, was used to identify pts diagnosed with stage IB-IIIA NSCLC without prior evidence of BM between Jan 2010 – December 2021. Presence of BM was identified from human curation of patient documents. Gradient boosting (GB), random forest (RF), and logistic regression (LR) algorithms with 3-fold cross-validation were trained and compared to predict risk of BM at two landmarks: 18 and 24 months (mos) from initial diagnosis (index). Pts who did not develop BM and either died or were lost to follow-up prior to landmark times were removed. Feature importance was defined using Shapley Additive Explanation (SHAP)-derived marginal odds ratios (OR). Due to low prevalence of BM, AUPRC was used as performance metric. Pts in the 1st and 4th quartiles of predicted probabilities from GB model were flagged as low-risk vs. high-risk. Results: Among 7473 pts in 18 mos model, median age was 68.4 years (IQR 13.2), 50.5% were female, and 10.9% were black. Demographics were similar for 6863 pts in 24 mos model. 6.4% and 8.3% developed BM at 18 and 24 mos. Ability of GB, RF, and LR models to predict BM was similar with validation AUPRC of 0.109 at 18 mos and 0.137 at 24 mos. In the GB model, BM prevalence in high-risk vs. low-risk group was 10.3% vs. 3.1% at 18 mos and 13.4% vs. 4.4% at 24 mos. In both landmark models, N0 stage and surgery within 90 days after index diagnosis were protective against BM, while presence of EGFR targetable mutations, adenocarcinoma (AD) histology, higher platelets (PLT), and history of pneumonia were risk factors. A glucose-by-histology interaction was found: For pts with normal blood glucose (GLU), risk of BM was independent of histology, for pts with high GLU, AD conferred greater risk of BM. Conclusions: An EHR-based ML model identified risk factors for developing BM among pts with eNSCLC and may identify pts who would benefit from active brain MRI surveillance and treatment intensification. [Table: see text]

A disease registry study to prospectively observe treatment patterns and outcomes in patients with HER2+ unresectable locally advanced or metastatic breast cancer (aBC): Interim results of the ESTHER study (NCT02393924).

e13039 Background: Over the last decade the standard-of-care treatment for patients with HER2+ aBC has included anti-HER2 therapies often combined with chemotherapy. These approaches have shown efficacy in clinical trials which have recruited patients in restricted populations defined by lines of therapy and previous treatments. There are few data regarding the long-term outcomes of patients treated with these regimens, including those with brain metastases. We conducted a prospective, multicentre non-interventional study to observe anti-cancer treatment regimens and clinical outcomes in patients with HER2 positive aBC. Methods: Adult patients diagnosed with HER2-positive aBC within the last 6 months were included. No specific treatment approach or course was mandated by the protocol. Source data was taken from the patient’s chart and other medical records, then reported by means of an electronic data collection system. Data regarding treatment received, efficacy (progression-free and overall survival (PFS), response rates (RR) and development of CNS metastases) and safety (including cardiac toxicity) were captured. Results: Between February 2015 and April 2018, 311 eligible patients were recruited with median age 57 years (24-96). At diagnosis of aBC the most common distant metastatic sites were bone (n = 145;47%), liver (n=117;38%), lung (n=100;32%) and brain (n=21;7%). First line treatment was with pertuzumab (P)/trastuzumab(T)/chemotherapy (n=212;68%), T/chemotherapy (n=45;14%), other T combinations without P (n=19;6%), Trastuzumab Emtansine regimens without T or P (n=16;5%) (19; 5% other combinations). Patients’ status as of cut-off date for this analysis (28th Feb 2022, median follow-up 44.1 months) is shown. The median PFS of patients treated in the 1st, 2nd and 3rd lines was 26.7 months, 10.7 months and 8.6 months. At the time of this analysis an additional 60 patients (19%) had developed brain metastases during first line. The number of patients starting 2nd, 3rd and 4th line treatment with brain metastases were 28 (22%), 17 (40%), 7 (37%) respectively. Conclusions: In routine clinical practice, as reflected by this registry, at least 40% of patients who initiate first line therapy for advanced breast cancer do not proceed to 3rd line therapy, and brain metastases become increasingly common. These findings have implications for selection of optimal 2nd and 3rd line treatment. Clinical trial information: ISRCTN02393924 . [Table: see text]

Multi-cancer brain metastasis risk score development and validation using 220,246 whole transcriptomes and machine learning.

2039 Background: Brain metastases occur in multiple cancer types with higher prevalence in lung, breast, melanoma, and GI cancers. The prognoses of patients who develop brain metastases are very poor and identification of brain metastasis risk could be useful for prognostication, monitoring, and therapy selection. Methods: Data from the whole transcriptome of 220,246 tumor profiles were analyzed and multiple machine learning models were trained on various molecular subtypes. The dataset was split 50% for training and the other 50% for testing, UMAP was employed for dimensionality reduction and the patterns learned across the entirety of the training dataset irrespective of brain metastasis were leveraged on the testing data set. Patients with brain metastasis were identified using the presence of ICD-10 code C79.31 (Secondary malignant neoplasm of the brain). As the absence of C79.31 could be due to the event not happening yet, patients without brain metastasis were stratified into groups based on 3, 4 or greater than 5 years without a C79.31 ICD-10 code. The brain metastasis risk score was defined by empirical evaluation of the positive predictive value in 7 groups of risk probabilities. The validation set contained 1,217 patients with brain metastasis and 4,631 without an observed brain metastasis within 3 years. Results: In the validation set, the prevalence of brain metastases within the risk scores across all cancer types ranged from 4% with the lowest risk score to 94% in the highest with 71% of cases receiving the lowest 2 risk scores, 15% the 2 intermediate risk scores, and 14% the 3 highest risk scores. For breast, lung and colon cancers, the prevalence of brain metastasis ranged from 4-10% in patients with the lowest risk scores to 92-100% in patients with the highest however the distribution of cases with each risk score was markedly different across cancer type. Breast cancer had 62% of cases receiving the lowest 2 risk scores versus 27% in lung, and 92% in colon. Breast cancer had 18% of cases receiving the 3 highest risk scores while lung had 42% and colon only 2% of cases with those 3 highest scores. Conclusions: Whole transcriptome data can be leveraged by a machine learning platform that employs dimensionality reduction techniques along with transfer learning to predict the risk of brain metastasis. This tool can be used to augment the clinical picture of cancer patients an unmet clinical opportunity to inform prognosis, monitoring, and therapeutic selection.

Phase II study of nivolumab (nivo) in combination with relatlimab (rela) in patients (pts) with active melanoma brain metastases (MBM).

TPS9605 Background: More than half of all pts diagnosed with metastatic melanoma (MM) will develop brain metastases. Despite this prevalence, pts with MBM are excluded from the majority of clinical trials, preventing them from potential benefit of novel therapies. Long-term outcomes of the CheckMate 204 study combining ipilimumab (ipi) (3mg/kg) plus nivo (1 mg/kg) for asymptomatic MBM reported an intracranial (IC) response in 54% of pts. The 36-month IC progression-free survival (PFS) was 54% and overall survival (OS) was 72%. Grade 3 or 4 treatment-related adverse events (TRAEs) occurred in 55% of pts. RELATIVITY-047, a global phase II/III randomized study, reported on the fixed-dose combination of nivo (480 mg) and rela (a LAG-3–blocking antibody,160 mg) compared to nivo (480 mg) alone for pts with advanced unresectable and untreated melanoma. The combination demonstrated significantly improved PFS and improved OS and objective response rate (ORR) compared to monotherapy. At 19.3 months of median follow-up, only the PFS benefit was statistically significant with a hazard ratio (HR) of 0.78 (95% CI 0.64-0.94). The OS benefit was not statistically significant (HR 0.8, 95% CI 0.64-1.01), and the descriptive ORR was 43.1% compared to 32.6% for single agent nivo. Grade 3 or 4 TRAEs occurred in 21% of pts in the combination group compared to 11% in the monotherapy group with no new safety signals. The nivo/rela combination has been FDA approved for the treatment of pts with MM but has not been studied in pts with MBM. We hypothesize that nivo/rela will be safe and tolerable for pts with MBM and demonstrate favorable outcomes compared to nivo monotherapy with fewer AEs in contrast to ipi/nivo combination. Methods: This single arm, single center, open-label phase II trial will evaluate the safety and efficacy of nivo/rela in pts with MBM (NCT05704647). 30 MBM pts who are treatment naïve to anti-PD-1 therapy will be enrolled. At least one unirradiated parenchymal brain metastasis at least 5 mm or larger is required for inclusion. Asymptomatic pts not requiring steroids for the management of their neurological symptoms are eligible. Pts requiring urgent local therapy with radiation or surgery are not eligible. Pts will be treated with nivo (480 mg)/rela (160 mg) every 4 weeks for up to 25 cycles, disease progression, or unacceptable toxicity. The primary objective is to evaluate IC ORR (CR + PR) by MRI per modified RECIST 1.1 criteria. Monitoring for futility will be performed using the Bayesian Optimal Phase 2 (BOP2) design. Thirty evaluable patients will provide 82% power to detect a difference between the historical IC ORR of 20% for nivo alone to a targeted ORR of 40% for nivo/rela. Longitudinal research blood, tissue, and microbiome samples will be collected along with neurocognitive assessment and quality of life surveys. Clinical trial information: NCT05704647 .

Characteristics of long-term survivors with EGFR mutant (EGFRm) metastatic non-small cell lung cancer (mNSCLC).

9116 Background: Although osimertinib has become standard first-line (1L) therapy for patients (pts) with EGFRm mNSCLC, a subset treated with earlier-generation EGFR-targeted TKIs and chemotherapy have had long-term survival. We sought to characterize clinical features of long-term survival in pts with EGFRm mNSCLC treated prior to the osimertinib era. Methods: Data were abstracted from electronic medical records at 12 cancer centers participating in the Academic Thoracic Oncology Medical Investigator’s Consortium (ATOMIC). We included patients with mNSCLC with sensitizing mutations in EGFR who started 1L systemic therapy before 2015. Survival distributions and predictors were assessed using Kaplan-Meier estimates and a multivariable Cox proportional hazards model including time-dependent brain metastasis (met) development, age, 1L therapy (targeted vs. chemotherapy), sex, race, and smoking status. Multivariable logistic regression was used to evaluate baseline predictors associated with 5+ year survival vs. not for those with known survival status at 5 years after start of 1L. Results: We identified 304 patients (69% female, 56% White, 29% Asian, mean age 61.2). First-line targeted therapy was given in 70% of patients. With a median follow-up of 81.5 months, median overall survival was 63.5 months (95% CI 59.4-71.9). 135 (44%) pts survived 5+ years; among those with baseline next-generation sequencing (NGS), presence of a baseline pathogenic ERBB2 variant was higher in 5+ year survivors (4/51 [8%], 2x amplification, y772_A775dup, S310F) v others (1/65 [2%], amplification). Among 161 pts with baseline brain imaging, both baseline and on-treatment development of brain mets were associated with worse survival (HR 1.29, 95% CI 1.21-1.37; HR 1.59, 95% CI 1.47-1.72; both P < 0.001). Excluding 22 patients lost to follow-up before 5 years, a history of smoking (OR 2.99, 95% CI 1.35-6.90, P = 0.008) and baseline brain metastases (OR 3.57, 95% CI 1.64-8.13, P = 0.002) were associated with death before 5 years. Conclusions: Long-term survivors with EGFRm mNSCLC treated before the osimertinib era were more likely to be nonsmokers and have no baseline brain metastases. This highlights a subset of EGFRm patients who had excellent outcomes with older treatment strategies and may not benefit as much from intensification approaches. Additional baseline mutational data will be presented at the conference.

A radiomic-based predictive model of lung adenocarcinoma brain metastases and molecular subtypes.

2014 Background: Non-small cell lung cancer (NSCLC) comprises the largest portion of brain metastases (BM) from solid cancer with 40% of patients developing BM during the course of their disease. There are currently no reliable prediction tools for identifying patients at risk for BM, especially in the early-stage setting where MRI screening is not performed. Furthermore, in the later stage setting, brain MRI are only performed annually. Therefore, there is a critical need to identify high-risk patients for BM that could benefit from MRI surveillance. Methods: We identified 162 lung adenocarcinoma (LUAD) patients with (N = 66) or without (N = 96) BM that had treatment-naïve CT scans with a segmentable lesion. The tumor, surrounding ground glass opacity and necrosis were segmented via 3D slicer to create a volume of interest for radiomic texture analysis and 400 features were extracted. The Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression feature selection method was used to select the most relevant features and models were built using the machine learning method XGBoost classifier. Training and testing sets with random splitting was used for cross validation. We report the accuracy, sensitivity, specificity, and area under the curve (AUC) for each model. Results: Among the extracted features that LASSO deemed as most discriminative for development of BM, we identified the most relevant features using XGBoost that predicted BM with 79% accuracy, 83% sensitivity, 72% specificity, and 79% AUC ( p= 0.01) in the overall population. The addition of ground glass opacity and necrosis to the model did not significantly improve performance. Furthermore, the model distinguished those with metachronous vs synchronous BM with 84% accuracy, 83% sensitivity, 86% specificity, and 83% AUC ( p = 0.04). Our model held up across molecular subtypes ( EGFR and KRAS mutant). Importantly, the model was predictive in early-stage patients with 92% accuracy, 96% sensitivity, 83% specificity, and 95% AUC ( p= 0.0005). Moreover, our model predicted for high vs. low overall survival, and was BM-specific as it was not predictive of other sites of metastases. We further developed a model from the CT features that correctly classified KRAS mutant vs. KRAS wild type LUAD with 77% accuracy, 73% sensitivity, 80% specificity, and 80% AUC ( p= 0.002). Conclusions: Utilizing a radiomics approach, we were able to predict BM from primary lung CT features including in stage I and II disease, predict synchronous vs metachronous BM, and distinguish distinct molecular LUAD subtypes. We are currently validating our BM prediction model in a large independent cohort and developing models to classify targetable LUAD-BM molecular alterations utilizing brain MRI scans. These studies will identify patients that require MRI surveillance in the early-stage setting and more intensive surveillance in the late-stage setting for BM.

A radiomics model developed from pre-metastatic brain magnetic resonance imaging to predict brain metastasis in non–small-cell lung cancer: Evidence to the seed-and-soil theory.

2030 Background: Steven Paget’s “seed and soil” theory suggests that both the primary tumor and the target host organ play important roles in successful metastasis development. In this proof-of-concept study, we proposed that inter-individual differences exist in the brain’s congeniality for developing brain metastasis (BM), and validated the theory via developing a non-invasive radiomics BM prediction model. Methods: 256 non-small cell lung cancer (NSCLC) patients with no BM at baseline brain magnetic resonance imaging (MRI) were selected, 128 patients developed BM within three years after diagnosis and 128 remained BM-free. For radiomics analysis, both the BM and non-BM groups were randomly distributed into training and testing datasets at an 70%:30% ratio. Baseline brain MRI (representing the soil) and chest computed tomography (CT, representing the seed) radiomics features were extracted to develop the BM prediction models. We first developed radiomics models using the training dataset and subsequently validated the models in the testing dataset. A radiomics BM score (RadBM score) was generated and BM-free survival were compared between RadBM score-high and RadBM score-low groups. Results: The radiomics model developed from baseline brain MRI features alone can predict BM development in NSCLC patients. Furthermore, a fusion model integrating brain MRI features with primary tumor CT features (seed-and-soil model) provided synergetic effect and was more efficient in predicting BM (areas under the receiver operating characteristic curve 0.84[95% confidence interval: 0.80–0.89] and 0.80 [95% confidence interval: 0.71–0.88] in the training and testing datasets, respectively). BM-free survival was significantly shorter in the RadBM score-high group versus the RadBM score-low group (Log-rank P < 0.001). Cumulative BM rate at 3-year were 75.8% and 24.2% for RadBM score-high and RadBM score-low group, respectively. Conclusions: The results demonstrated that intrinsic features of a non-metastatic organ exert a significant impact on metastasis development, which is first-in-class in metastasis prediction studies and provided novel evidence for the "seed-and-soil" theory of tumor metastasis. Radiomics BM prediction model utilizing both pre-metastatic brain and primary tumor features might provide a useful tool for identifying NSCLC patients more prone to develop BM and providing individualized management for these patients.

The Association of Gross Tumor Volume and Its Radiomics Features with Brain Metastases Development in Patients with Radically Treated Stage III Non-Small Cell Lung Cancer.

To identify clinical risk factors, including gross tumor volume (GTV) and radiomics features, for developing brain metastases (BM) in patients with radically treated stage III non-small cell lung cancer (NSCLC). Clinical data and planning CT scans for thoracic radiotherapy were retrieved from patients with radically treated stage III NSCLC. Radiomics features were extracted from the GTV, primary lung tumor (GTVp), and involved lymph nodes (GTVn), separately. Competing risk analysis was used to develop models (clinical, radiomics, and combined model). LASSO regression was performed to select radiomics features and train models. Area under the receiver operating characteristic curves (AUC-ROC) and calibration were performed to assess the models' performance. Three-hundred-ten patients were eligible and 52 (16.8%) developed BM. Three clinical variables (age, NSCLC subtype, and GTVn) and five radiomics features from each radiomics model were significantly associated with BM. Radiomic features measuring tumor heterogeneity were the most relevant. The AUCs and calibration curves of the models showed that the GTVn radiomics model had the best performance (AUC: 0.74; 95% CI: 0.71-0.86; sensitivity: 84%; specificity: 61%; positive predictive value [PPV]: 29%; negative predictive value [NPV]: 95%; accuracy: 65%). Age, NSCLC subtype, and GTVn were significant risk factors for BM. GTVn radiomics features provided higher predictive value than GTVp and GTV for BM development. GTVp and GTVn should be separated in clinical and research practice.

A Survival Tree of Advanced Melanoma Patients with Brain Metastases Treated with Immune Checkpoint Inhibitors.

The efficacy of immune checkpoint inhibitors (ICIs) in patients with advanced melanoma that develop brain metastases (BM) remains unpredictable. In this study, we aimed to identify prognostic factors in patients with melanoma BM who are treated with ICIs. Data from advanced melanoma patients with BM treated with ICIs in any line between 2013 and 2020 were obtained from the Dutch Melanoma Treatment Registry. Patients were included from the time of the treatment of BM with ICIs. Survival tree analysis was performed with clinicopathological parameters as potential classifiers and overall survival (OS) as the response variable. In total, 1278 patients were included. Most patients were treated with ipilimumab-nivolumab combination therapy (45%). The survival tree analysis resulted in 31 subgroups. The median OS ranged from 2.7 months to 35.7 months. The strongest clinical parameter associated with survival in advanced melanoma patients with BM was the serum lactate dehydrogenase (LDH) level. Patients with elevated LDH levels and symptomatic BM had the worst prognosis. The clinicopathological classifiers identified in this study can contribute to optimizing clinical studies and can aid doctors in giving an indication of the patients' survival based on their baseline and disease characteristics.

Clinical characteristics, treatment, and long-term outcome of patients with brain metastases from thyroid cancer

Brain metastases (BM) in patients with thyroid cancer (TC) are rare with an incidence of 1% for papillary and follicular, 3% for medullary and up to 10% for anaplastic TC (PTC, FTC, MTC and ATC). Little is known about the characteristics and management of BM from TC. Thus, we retrospectively analyzed patients with histologically verified TC and radiologically verified BM identified from the Vienna Brain Metastasis Registry. A total of 20/6074 patients included in the database since 1986 had BM from TC and 13/20 were female. Ten patients had FTC, 8 PTC, one MTC and one ATC. The median age at diagnosis of BM was 68 years. All but one had symptomatic BM and 13/20 patients had a singular BM. Synchronous BM at primary diagnosis were found in 6 patients, while the median time to BM diagnosis was 13 years for PTC (range 1.9–24), 4 years for FTC (range 2.1–41) and 22 years for the MTC patient. The overall survival from diagnosis of BM was 13 months for PTC (range 1.8–57), 26 months for FTC (range 3.9–188), 12 years for the MTC and 3 months for the ATC patient. In conclusion, development of BM from TC is exceedingly rare and the most common presentation is a symptomatic single lesion. While BM generally constitute a poor prognostic factor, individual patients experience long-term survival following local therapy.

Cognitive adverse events in patients with lung cancer treated with checkpoint inhibitor monotherapy: a propensity score-matched analysis.

Cancer-related cognitive decline is a serious problem in long-term survival but no pivotal study has investigated whether checkpoint inhibitors (ICI) may be associated with cognitive adverse events. This propensity score-matched analysis recruited non-small cell lung cancer (NSCLC) patients prescribed with or without ICI monotherapy from three Chinese tertiary hospitals. Patients were excluded from study who developed brain metastasis or had disorders severely affecting cognitive abilities. Primary outcomes were changes in neuropsychological battery test (NBT) at baseline, 6- and 12-month sessions, and any NBT score changes that exceeded 3∗SD of baseline scores would be marked as objective cognitive adverse events (CoAE). Secondary endpoint was the 20-item Perceived Cognitive Impairment (PCI) sub-scale score change in Functional Assessment of Cancer Therapy-Cognitive Function questionnaire, administered at baseline, 3-, 6-, 9-, 12-, and 15-month follow-up session. Per-protocol ICI and control arms were matched with propensity scores that incorporated baseline variables to compare both NBT and PCI assessment results. Patients participating in PCI assessments were analysed in intention-to-treat analysis. Kaplan-Meier survival curves with log-rank tests were adopted to analyse incidence of perceived cognitive decline events (PCDE). Between March 12, 2020, and March 28, 2021, 908 participants were enrolled. Compared to control, 3 of 4 subtest of NBT scores in ICI arm showed significant cognitive decline in 6- and 12-month sessions, in which Trail Making Test score change (13.56±11.73) reached threshold of cognitive deficit diagnosis in the 12-month session. In 1:1 matched 292 pairs from 908 patients, PCI score changes in ICI arms were-4.26±8.54 (3rd month),-4.72±11.83 (6th month),-6.16±15.41 (9th month),-6.07±15.71 (12th month), and-7.96±13.97 (15th month). The scores were significantly lower than control arm in 3-, 6-, and 12-session follow-up. The result was validated after adjusting quality of life scores and in intention-to-treat analysis. Mean PCI change exceeded 1/2 SD of baseline PCI score (5.81) in 9-, 12-, and 15-month sessions in ICI arm, but not in control arm. PCDE incidence/prevalence was significantly higher in ICI arm (incidence 26.4% vs. 5.1%, and prevalence 16.2% vs. 1.7%). Immune-related adverse events related to incidence of PCDE after adjusting for baseline variables. ICI monotherapy seemed to relate to higher cognitive decline represented by score changes and incidence/prevalence rates. The decline deteriorated as treatment progressed, and immune-related adverse events seemed to be associated with higher cognitive adverse events incidence in the ICI treatment. The Fellowship of China Postdoctoral Science Foundation and National Natural Science Foundation of China Youth Science Fund Project.