Deep learning-based detection and quantification of brain metastases on black-blood imaging can provide treatment suggestions: a clinical cohort study.

OBJECTIVES: We aimed to evaluate whether deep learning-based detection and quantification of brain metastasis (BM) may suggest treatment options for patients with BMs. METHODS: The deep learning system (DLS) for detection and quantification of BM was developed in 193 patients and applied to 112 patients that were newly detected on black-blood contrast-enhanced T1-weighted imaging. Patients were assigned to one of 3 treatment suggestion groups according to the European Association of Neuro-Oncology (EANO)-European Society for Medical Oncology (ESMO) recommendations using number and volume of the BMs detected by the DLS: short-term imaging follow-up without treatment (group A), surgery or stereotactic radiosurgery (limited BM, group B), or whole-brain radiotherapy or systemic chemotherapy (extensive BM, group C). The concordance between the DLS-based groups and clinical decisions was analyzed with or without consideration of targeted agents. The performance of distinguishing high-risk (B + C) was calculated. RESULTS: Among 112 patients (mean age 64.3 years, 63 men), group C had the largest number and volume of BM, followed by group B (4.4 and 851.6 mm3) and A (1.5 and 15.5 mm3). The DLS-based groups were concordant with the actual clinical decisions, with an accuracy of 76.8% (86 of 112). Modified accuracy considering targeted agents was 81.3% (91 of 112). The DLS showed 95% (82/86) sensitivity and 81% (21/26) specificity for distinguishing the high risk. CONCLUSION: DLS-based detection and quantification of BM have the potential to be helpful in the determination of treatment options for both low- and high-risk groups of limited and extensive BMs. CLINICAL RELEVANCE STATEMENT: For patients with newly diagnosed brain metastasis, deep learning-based detection and quantification may be used in clinical settings where prompt and accurate treatment decisions are required, which can lead to better patient outcomes. KEY POINTS: • Deep learning-based brain metastasis detection and quantification showed excellent agreement with ground-truth classifications. • By setting an algorithm to suggest treatment based on the number and volume of brain metastases detected by the deep learning system, the concordance was 81.3%. • When dividing patients into low- and high-risk groups, the sensitivity for detecting the latter was 95%.

Vessel size and perfusion-derived vascular habitat refines prediction of treatment failure to bevacizumab in recurrent glioblastomas: validation in a prospective cohort.

OBJECTIVES: Anti-angiogenic therapy may not benefit all patients with recurrent glioblastomas, and imaging biomarker predicting treatment response to anti-angiogenic therapy is currently limited. We aimed to develop and validate vascular habitats based on perfusion and vessel size to predict time to progression (TTP) in patients with recurrent glioblastomas treated with bevacizumab. METHODS: Sixty-nine patients with recurrent glioblastomas treated with bevacizumab who underwent pretreatment MRI with dynamic susceptibility contrast imaging and vessel architectural imaging were enrolled. Vascular habitats were constructed using vessel size index (VSI) and relative cerebral blood volume (rCBV). Associations with vascular habitats and TTP were analyzed using Cox proportional hazard regression analysis. In a prospectively enrolled validation cohort consisting of 15 patients ( ClinicalTrials.gov identifier; NCT04143425), stratification of TTP was demonstrated by the Kaplan-Meier method (log-rank test) using vascular habitats. RESULTS: Three vascular habitats consisting of high, intermediate, and low angiogenic habitats were identified with rCBV and VSI. Both high angiogenic and intermediate angiogenic habitats were significantly associated with a shorter TTP (hazard ratio [HR], 2.78 and 1.82, respectively; largest p = .003) and so was rCBV (HR, 2.15; p = .02). Concordance probability index of vascular habitat combining high and intermediate angiogenic habitats was 0.74. Vascular habitats stratified patients as good or poor responder in a prospective cohort (p = .059). CONCLUSIONS: Perfusion- and vessel size-derived vascular habitats predicted TTP in recurrent glioblastomas treated with anti-angiogenic therapy and aided patient stratification in a prospective validation cohort. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT04143425 KEY POINTS: • High and intermediate angiogenic habitats predicted TTP in recurrent glioblastomas treated with anti-angiogenic therapy. • Vascular habitats combining high and intermediate angiogenic habitats aided patient stratification for anti-angiogenic therapy in recurrent glioblastomas.

Clinical implications of the tumor microenvironment using multiplexed immunohistochemistry in patients with advanced or metastatic renal cell carcinoma treated with nivolumab plus ipilimumab.

Purpose: Immune checkpoint inhibitors (ICIs) such as nivolumab and ipilimumab (N/I) are important treatment options for advanced renal cell carcinoma (RCC). The tumor microenvironment (TME) in these ICI-treated patients is largely unknown. Methods: Twenty-four patients treated with N/I between July 2015 and June 2020 were analyzed. Multiplexed immunohistochemistry (mIHC) was conducted to define the TME, including various T cell subsets, B cells, macrophages, and dendritic cells. Results: The median age of the study patients was 61 years (range, 39-80) and 75.0% of these cases were men. The objective response rate with N/I was 50.0%. The densities of the CD8+ cytotoxic T cells (P=0.005), specifically CD137+ CD8+ T cells (P=0.017), Foxp3- CD4+ helper T cells (P=0.003), Foxp3+ CD4+ regulatory T cells (P=0.045), CD68+ CD206- M1 macrophages (P=0.008), and CD68+ CD206+ M2 macrophages (P=0.021) were significantly higher in the treatment responders. At a median follow-up duration of 24.7 months, the median progression-free survival (PFS) was 11.6 months. The high densities (≥median) of Foxp3- CD4+ helper T cells (P=0.016) and CD68+ CD206- M1 macrophages (P=0.008) were significantly associated with better PFS, and the density of CD137+ CD8+ cytotoxic T cells (P=0.079) was marginally associated with better PFS. After multivariate analysis, the higher density of Foxp3- CD4+ helper T cells was independently associated with better PFS (hazard ratio 0.19; P=0.016). Conclusion: The properties and clinical implications of the TME properties in RCC indicate that Foxp3- CD4+ helper T cells, M1 macrophages, and CD137+ CD8+ T cells are potential predictive biomarkers and treatment targets.

Clinical course of patients with renal cell carcinoma or urothelial carcinoma who had stable disease as an initial response to a PD-1 or PD-L1 inhibitor.

BACKGROUND: About 17%-34% of patients with non-prostatic genitourinary (GU) cancer had stable disease (SD) as their best response to programmed cell death protein 1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors. However, most efficacy studies of PD-1/PD-L1 inhibitors are focused on the response and its durability. Little is known about patients' clinical outcomes with SD as their initial response to PD-1/PD-L1 inhibitors. METHODS: We retrospectively reviewed the clinical outcomes of patients with renal cell carcinoma (RCC) (n = 102) and urothelial carcinoma (UC) (n = 101) treated with PD-1/PD-L1 inhibitors. The duration of SD (DoSD) was calculated for patients who had SD as their best response and was defined from the time of SD to radiologically confirmed progressive disease (PD) or death. RESULTS: Fifty-five patients (27.1%) had SD as the initial response. Among them, 10 patients (18.2%) achieved a response on subsequent assessments. With a median follow-up duration of 15.2 months, the median DoSD was 8.5 months (range: 3.2-13.8), which was significantly different according to the disease; 9.4 months in RCC and 2.3 months in UC (p = 0.03). If the disease did not progress in 4 months (long SD subgroup), the DoSD (11.0 months) was comparable to the duration of response (12.9 months) in patients who achieved a complete or partial response to PD-1/PD-L1 inhibitors. In addition to the cancer type, the clinical factors associated with short DoSD were liver metastases (p = 0.003), neutrophil-lymphocyte ratio (NLR) > 4.0 (p = 0.001), and platelet-lymphocyte ratio (PLR) > 194 (p = 0.003). CONCLUSION: For RCC patients with SD to PD-1/PD-L1 inhibitors, if there are no liver metastases and NLR/PLR is below the certain level, they would have similar duration of disease control and survival benefit as those who achieve the response.

Use of Gemcitabine plus Carboplatin is Associated with Poor Outcomes in Urothelial Carcinoma Patients with Chronic Kidney Disease Stage 4-5.

PURPOSE: This study aimed to investigate the clinical outcomes with gemcitabine-carboplatin (GCb), the standard treatment for patients with advanced urothelial carcinoma (UC) who are ineligible for cisplatin-based regimens, in advanced UC patients with a glomerular filtration rate (GFR) < 30 mL/min. MATERIALS AND METHODS: A retrospective cohort study involving GCb-treated advanced UC patients with GFR < 60 mL/min (n=89) was performed. Clinical outcomes were compared between subgroups with GFR < 30 mL/min and GFR ≥ 30 mL/min but < 60 mL/min. RESULTS: Most baseline characteristics were comparable between the two subgroups. Patients with GFR < 30 mL/min had a significantly lower objective response rate (12.5%) compared to those with higher GFR levels (56.7%) (p=0.004). The number of GCb cycles was significantly lower in patients with GFR < 30 mL/min (median 2 cycles) than in those with higher GFR levels (median 6 cycles) (p=0.002). Compared to those with GFR ≥ 30 mL/min but < 60 mL/min, patients with GFR < 30 mL/min showed significantly worse progression-free survival (PFS) and overall survival (OS) (p < 0.001 for both). Further stratification of patient subgroups according to their GFR (i.e., GFR ≥ 45 mL/min but < 60 mL/min vs. GFR ≥ 30 mL/min but < 45 mL/min vs. GFR < 30 mL/min) revealed significantly different PFS and OS (p < 0.001 for both). CONCLUSION: The use of GCb is discouraged in advanced UC patients with GFR < 30 mL/min. Alternative therapeutic approaches with better efficacy are warranted for these patients.

Radiomics-based prediction model for outcomes of PD-1/PD-L1 immunotherapy in metastatic urothelial carcinoma.

OBJECTIVES: To evaluate the usefulness of a radiomics-based prediction model for predicting response and survival outcomes of patients with metastatic urothelial carcinoma treated with immunotherapy targeting programmed cell death 1 (PD-1) and its ligand (PD-L1). METHODS: Sixty-two patients who underwent immunotherapy were divided into training (n = 41) and validation sets (n = 21). A total of 224 measurable lesions were identified on contrast-enhanced CT. A radiomics signature was constructed with features selected using a least absolute shrinkage and selection operator algorithm in the training set. A radiomics-based model was built based on a radiomics signature consisting of five reliable RFs and the presence of visceral organ involvement using multivariate logistic regression. According to a cutoff determined on the training set, patients in the validation set were assigned to either high- or low-risk groups. Kaplan-Meier analysis was performed to compare progression-free and overall survival between high- and low-risk groups. RESULTS: For predicting objective response and disease control, the areas under the receiver operating characteristic curves of the radiomics-based model were 0.87 (95% CI, 0.65-0.97) and 0.88 (95% CI, 0.67-0.98) for the validation set, providing larger net benefit determined by decision curve analysis than without radiomics-based model. The high-risk group in the validation set showed shorter progression-free and overall survival than the low-risk group (log-rank p = 0.044 and p = 0.035). CONCLUSIONS: The radiomics-based model may predict the response and survival outcome in patients treated with PD-1/PD-L1 immunotherapy for metastatic urothelial carcinoma. This approach may provide important and decision tool for planning immunotherapy. KEY POINTS: • A radiomics-based model was built based on radiomics features and the presence of visceral organ involvement for prediction of outcomes in metastatic urothelial carcinoma treated with immunotherapy. • This prediction model demonstrated good prediction of treatment response and higher net benefit than no model in the independent validation set. • This radiomics-based model demonstrated significant associations with progression-free and overall survival between low-risk and high-risk groups.

Impact of clinicopathologic features on leptomeningeal metastasis from lung adenocarcinoma and treatment efficacy with epidermal growth factor receptor tyrosine kinase inhibitor.

BACKGROUND: We investigated the risk factors for leptomeningeal carcinomatosis (LMC) and compared clinical efficacies of various treatment modalities including intrathecal (IT) chemotherapy in patients with lung adenocarcinoma harboring epidermal growth factor receptor (EGFR) mutations. METHODS: Using clinical research data from the Asan Medical Center, we retrospectively analyzed data of patients diagnosed with LMC, confirmed via cerebrospinal fluid (CSF) analysis from January 2008 to December 2017. RESULTS: We identified 1189 patients with lung adenocarcinoma harboring EGFR mutations. Among these, 9.8% had a median duration of 13.5 (interquartile range [IQR] 6.8-23.6) months from the initial lung cancer diagnosis to LMC occurrence. Younger age (hazard ratio [HR] 1.043, P < 0.001), initial metastatic disease (HR 3.768, P < 0.001), and metastasis to the brain (HR 8.682, P < 0.001) or lung (HR 2.317, P = 0.004) were risk factors associated with LMC. Median survival duration from LMC diagnosis was 3.8 (IQR 1.5-8.6) months. Eastern Cooperative Oncology Group performance status score ≤ 2 (HR 0.505, P = 0.007) and insertion of Ommaya reservoir (HR 0.445, P = 0.005) were associated with longer survival. EGFR-tyrosine kinase inhibitor (TKI) conferred survival benefits compared to cytotoxic chemotherapy or best supportive care (HR 2.222, P = 0.018; HR 5.638, P < 0.001, respectively). Although IT chemotherapy showed no survival benefit, it was associated with improved neurologic symptoms and signs and CSF negative conversion. CONCLUSIONS: Younger age, initial diagnosis of metastatic disease, and metastasis to the brain or different lobes were associated with LMC in patients with EGFR-mutant lung adenocarcinoma. Therapeutic interventions including EGFR-TKIs, cytotoxic chemotherapy, or Ommaya reservoir, and good performance status were related to favorable survival outcomes. KEY POINTS: Age and disease status were associated with LMC in patients with EGFR-mutant adenocarcinoma, and EGFR-TKI, Ommaya reservoir, and good performance status were related to survival benefit.

How Cancer Patients Perceive Clinical Trials (CTs) in the Era of CTs: Current Perception and Its Differences Between Common and Rare Cancers.

Perception has recently been highlighted as a critical determinant for participation in clinical trials (CTs) among cancer patients. We evaluated cancer patients' current perceptions of CTs using the PARTAKE questionnaires, focusing on differences between patients with common and rare cancers. From November 2015 to May 2017, we prospectively surveyed patients who had received anti-cancer treatment at Asan Medical Center. Among 333 respondents, 70.9% had common and 29.1% had rare cancers. In the cohort, 87.7% of patients with common cancers and 75.3% of patients with rare cancers answered that they heard of and knew about CTs. However, willingness to participate in CTs was expressed only in approximately 56% of patients, although it was significantly associated with awareness and perception. Surprisingly, patients with rare cancers when compared with patients with common cancers showed significantly lower levels of awareness and perception (64.2% vs 79.9%, p = 0.003 and 77.3% vs 91.9%, p < 0.001), and consequently less willingness to participate in CTs (47.4% vs 58.9%, p = 0.06). In addition, cancer patients still harbored fear and concerns about safety and reward of CTs, and demonstrated substantial lack of knowledge about the voluntary nature of CTs, which was more obvious in patients with rare cancers. We identified relatively modest willingness of cancer patients to participate in CTs regardless of generally favorable perception. These findings are highlighted by the more negative perception of CTs among patients with rare cancers relative to those with common cancers. Further education and encouragement by research and public entities seem essential to improve motivation of CTs in cancer patients beyond good perception, especially for patients with rare cancers.

Hyperprogressive Disease in Patients With Urothelial Carcinoma or Renal Cell Carcinoma Treated With PD-1/PD-L1 Inhibitors.

BACKGROUND: A rapid progression pattern called hyperprogressive disease (HPD) has been observed during early cycles of programmed cell death protein 1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitor therapy. Data regarding HPD in patients with genitourinary cancer are limited. PATIENTS AND METHODS: We included 203 patients with genitourinary cancer treated with PD-1/PD-L1 inhibitors between February 2015 and June 2018. HPD was defined as a greater than 50% increase in tumor burden, greater than 2-fold increase in tumor growth rate, or development of extensive (10 or more) new lesions. RESULTS: Patients (n = 102) with renal cell carcinoma (RCC) and patients (n = 101) with urothelial carcinoma (UC) were included. HPD was observed in 13 (6.4%) patients. The median overall survival for patients with progressive disease and HPD was 7.3 months and 3.5 months, respectively. HPD occurred more frequently in patients with UC than in those with RCC (11.9% vs. 0.9%; P = .01). Multivariate analysis showed that UC and creatinine above 1.2 mg/dL were independent predictive factors for HPD. A 30% increase in lymphocyte number following PD-1/PD-L1 inhibitor treatment was a negative predictor of HPD. The incidence of HPD in patients with UC treated with paclitaxel-based chemotherapy was one-third of those treated with PD-1/PD-L1 inhibitors. CONCLUSION: HPD developed predominantly in patients with UC, and the incidence of HPD in patients with RCC was negligible. Treatment with PD-1/PD-L1 inhibitors should be prescribed with caution in patients with UC and creatinine above 1.2 mg/dL.

Efficacy, safety, and resistance profile of osimertinib in T790M mutation-positive non-small cell lung cancer in real-world practice.

The efficacy and safety of osimertinib were demonstrated in clinical trials; however, real-world clinical data, particularly the resistance profile, are limited. Here, we investigated the efficacy, safety, and resistance profile of osimertinib in real-world practice. We reviewed medical records of T790M mutation-positive lung cancer patients who started osimertinib between February 2016 and June 2017. Molecular pathologic data of biopsy samples obtained after acquisition of resistance to osimertinib were also analyzed. The study included 23 patients with a median age of 59 years. The median follow-up duration was 11.9 months (IQR, 4.7-15.8). Objective response was achieved in 17 (73.9%) patients, and the disease was controlled in 22 (95.7%) patients. Median progression-free survival (PFS) was 7.4 months (95% CI, 3.6-11.0). Adverse events were minimal except for one case of pneumonitis. Of 14 patients experiencing disease progression, 10 underwent re-biopsy. The T790M mutation disappeared in seven patients (70%), and one showed wild-type conversion. PFS was shorter in the T790M-loss group than in the T790M-persistent group (4.4 vs. 7.7 months). Two patients with small cell transformation responded well to subsequent chemotherapy. One patient developed a C797S mutation that became undetectable after two cycles of gemcitabine and cisplatin followed by six cycles of pembrolizumab, after which the patient responded well to osimertinib. In conclusion, osimertinib showed favorable efficacy and safety in real-world practice comparable to those observed in clinical trials. Repeat biopsy after the acquisition of resistance to osimertinib is helpful to direct further treatment strategies.

Prognostic implications of tumor extent in early-stage diffuse large B-cell lymphoma.

Therapeutic strategies for early-stage diffuse large B-cell lymphoma (DLBCL) are often influenced by tumor extent, but the prognostic value of this parameter is rarely defined. Here, a retrospective analysis was performed to define the impact of tumor extent on survival of patients with early-stage DLBCL. Eighty-six patients with stage II DLBCL, diagnosed from 2000-2007, were categorized into localized (n = 55, 64 %) and disseminated groups (n = 31, 36 %) based on tumor extent at time of diagnosis. Treatment modalities, chemotherapy regimen and number of chemotherapy cycles were the same between groups. With a median follow-up of 7.6 years (range 2.1-12.1 years), overall 5-year event-free survival (EFS) and overall survival (OS) were 70.6 and 76.5 %, respectively. EFS (P = 1.00) and OS (P = 0.20) did not differ between the two groups. Older age (>60 years) was significantly associated with poor EFS (P = 0.01) and OS (P = 0.04). High-risk patients as rated by stage-modified international prognostic index (IPI) had inferior EFS (P = 0.04) and OS (P = 0.06) compared with the intermediate-risk group. These results indicate that tumor extent has no prognostic value in patients with early-stage DLBCL. Consistent with previous studies, age and stage-modified IPI were useful prognostic indices for these patients.

Cross-sectional study of imatinib plasma trough levels in patients with advanced gastrointestinal stromal tumors: impact of gastrointestinal resection on exposure to imatinib.

PURPOSE: Imatinib plasma trough levels (C(min)) have been reported to correlate with treatment outcomes in patients with gastrointestinal stromal tumors (GISTs). We therefore have evaluated the correlation between imatinib C(min) and the clinical characteristics of patients with GIST. PATIENTS AND METHODS: Steady-state imatinib C(min) in 107 patients with GIST who were taking imatinib 300 to 800 mg/d was measured. RESULTS: In patients treated with imatinib 400 (n = 92), 300 (n = 7), 600 (n = 2), or 800 (n = 11) mg/d, imatinib C(min) was 1,305 +/- 633 ng/mL, 1,452 +/- 830 ng/mL, 1,698 +/- 725 ng/mL, and 3,330 +/- 1,592 ng/mL, respectively. Of the 92 patients treated with 400 mg/d imatinib, 59 patients (63%) were men; the median age was 55 years (range, 28 to 76 years), and the median duration of imatinib use before sampling was 8.8 months (range, 0.5 to 67.6 months). The mean inter- and intrapatient variability rates were 44.7% and 26.5%, respectively. In univariate analyses, high C(min) was correlated with advanced age (P = .02), low creatinine clearance (P = .001), low hemoglobin (P = .03), and low albumin (P < .001) concentrations. Imatinib C(min) was also significantly lower in patients with (n = 18; 942 +/- 330 ng/mL) than without (n = 74; 1,393 +/- 659 ng/mL) major (ie, total or subtotal) gastrectomy (P = .002). In multivariate analysis, albumin (P = .001) concentrations, creatinine clearance (P = .002), and major gastrectomy (P = .003) were significantly correlated with C(min). CONCLUSION: In patients with GIST, imatinib C(min) at steady state was significantly associated with albumin concentration, creatinine clearance, and previous major gastrectomy. Although its clinical impact is unclear at present time, monitoring of imatinib C(min) might be particularly important for optimal treatment with imatinib in patients who have undergone major gastrectomy.

Angiosarcoma of the retroperitoneum: report on a patient treated with sunitinib.

A 52 year-old woman presented with an incidentally detected retroperitoneal angiosarcoma and multiple hepatic metastases. After chemotherapy with weekly paclitaxel and doxorubicin, angiosarcoma had progressed rapidly. Because few chemotherapeutic options were available for her, sunitinib (37.5 mg/day, daily) as a salvage regimen was administered. Although sunitinib was interrupted after two weeks due to hematologic abnormalities, some metastatic nodules were regressed. Therefore, sunitinib was recommenced at a reduced dose (25 mg/day, daily). Serial computed tomography scans showed variable response in each tumor, however, sunitinib at least delayed tumor progression, compared to previous chemotherapy. With this case report, we suggest sunitinib may be effective against angiosarcomas. When sunitinib is administered to patients with angiosarcomas, hematologic abnormalities should be monitored frequently as severe hematologic toxicity may be caused either by sunitinib per se or angiosarcoma.