SUVmax Predicts Disease Progression after Stereotactic Ablative Radiotherapy in Stage I Non-small Cell Lung Cancer

Enhanced NSCLC subtyping and staging through attention-augmented multi-task deep learning: A novel diagnostic tool

OBJECTIVES

The objective of this study is to develop a novel multi-task learning approach with attention encoders for classifying histologic subtypes and clinical stages of non-small cell lung cancer (NSCLC), with superior performance compared to currently popular deep-learning models.

MATERIAL AND METHODS

Data were collected from six publicly available datasets in The Cancer Imaging Archive (TCIA). Following the inclusion and exclusion criteria, a total of 4548 CT slices from 758 cases were allocated. We evaluated multiple multi-task learning models that integrate attention mechanisms to resolve challenges in NSCLC subtype classification and clinical staging. These models utilized convolution-based modules in their shared layers for feature extraction, while the task layers were dedicated to histological subtype classification and staging. Each branch sequentially processed features through convolution-based and attention-based modules prior to classification.

RESULTS

Our study evaluated 758 NSCLC patients (mean age, 66.2 years ± 10.3; 473 men), spanning ADC and SCC cases. In the classification of histological subtypes and clinical staging of NSCLC, the MobileNet-based multi-task learning model enhanced with attention mechanisms (MN-MTL-A) demonstrated superior performance, achieving Area Under the Curve (AUC) scores of 0.963 (95 % CI: 0.943, 0.981) and 0.966 (95 % CI: 0.945, 0.982) for each task, respectively. The model significantly surpassed its counterparts lacking attention mechanisms and those configured for single-task learning, as evidenced by P-values of 0.01 or less for both tasks, according to DeLong's test.

CONCLUSIONS

The integration of attention encoder blocks into our multi-task learning network significantly enhanced the accuracy of NSCLC histological subtyping and clinical staging. Given the reduced reliance on precise radiologist annotation, our proposed model shows promising potential for clinical application.

Invasive Nodal Staging via Endobronchial Ultrasound and Outcome in Patients Treated with Stereotactic Body Radiation Therapy for Early-Stage Non-Small Cell Lung Cancer - Results from a Single Institution Study

INTRODUCTION

Stereotactic body radiation therapy (SBRT) is an effective treatment for medically inoperable early-stage non-small cell lung cancer (NSCLC). The prognostic value of invasive nodal staging (INS) for patients undergoing SRBT has not been studied extensively. Herein, we report the impact of INS in addition to 18F-FDG-PET on treatment outcome for patients with NSCLC undergoing SBRT.

MATERIALS AND METHODS

Patients with stage I/ II NSCLC who underwent SBRT were included with IRB approval. Clinical, dosimetric, and radiological data were obtained. Overall survival (OS), regional recurrence free survival (RRFS), local recurrence free survival (LRFS), and distant recurrence free survival (DRFS) were analyzed using Kaplan Meyer method. Univariable analysis (UVA) and multivariable analysis (MVA) were performed to assess the relationship between the variables and the outcomes.

RESULTS

A total of 376 patients were included in the analysis. Median follow up was 43 months (IQ 32.6-45.8). Median OS, LRFS, RRFS, DRFS were 40, 32, 32, 33 months, respectively. The 5-year local, regional, and distant failure rates were 13.4%, 23.5% and 25.3%, respectively. The 1-year, 3-year and 5-year OS were 83.8%, 55.6%, and 36.3%, respectively. On MVA, INS was not a predictor of either improved overall or any recurrence free survival endpoints while larger tumor size, age, and adjusted Charleston co-morbidity index (aCCI) were significant for inferior LRFS, RRFS, and DRFS.

CONCLUSION

Invasive nodal staging did not improve overall or recurrence free survival among patients with early-stage NSCLC treated with SBRT whereas older age, aCCI, and larger tumor size were significant predictors of LRFS, RRFS, and DRFS.

Tumor to liver maximum standardized uptake value ratio of FDG-PET/CT parameters predicts tumor treatment response and survival of stage III non-small cell lung cancer

BACKGROUND

To assess the predictive values of primary tumor FDG uptake for patients with inoperable stage III non-small cell lung cancer (NSCLC) after concurrent chemoradiotherapy (CCRT).

METHODS

A total of 107 patients with diagnosis of stage III NSCLC and CCRT were enrolled. The tumor maximum uptake value (SUVmax) was standardized by calculating several ratios between tumor and each background tissues. The receiver operating characteristics curve (ROC) was used to compare the predictive power of prognostic models. The tumor objective response rate (ORR) and overall survival (OS) were compared and analyzed by the Kaplan-Meier method and univariate and multivariate Cox regression models.

RESULTS

The areas under ROC curve (AUCs) ranged from 0.72 to 0.81 among these tumor SUVmax and standardized SUVmax ratios, and the tumor SUVmax and tumor SUVmax-to-liver SUVmean ratio (TLMR) were more predictive of ORR (AUC, 0.81; 95% CI, 0.73-0.88 for tumor SUVmax and AUC, 0.84; 95%CI, 0.76-0.91 for TLMR) than any of other SUVmax ratios. The patients with lower tumor SUVmax, SUVmean and SUVmax ratios had a significantly better OS than those with their corresponding higher ones. Moreover, both univariate and multivariable analyses revealed that TLMR was significantly associated with better ORR and OS after adjustment with other prognostic variables.

CONCLUSIONS

TLMR, a standardized tumor SUVmax, was an independent prognostic predictor for tumor ORR and OS of patients with stage III NSCLC after CCRT.

Prognostic value of consolidation-to-tumor ratio on computed tomography in NSCLC: a meta-analysis

BACKGROUND

Although several studies have confirmed the prognostic value of the consolidation to tumor ratio (CTR) in non-small cell lung cancer (NSCLC), there still remains controversial about it.

METHODS

We systematically searched the PubMed, Embase, and Web of Science databases from inception to April, 2022 for eligible studies that reported the correlation between CTR and prognosis in NSCLC. Hazard ratios (HRs) with 95% confidence intervals (95% CIs) were extracted and pooled to assess the overall effects. Heterogeneity was estimated by I² statistics. Subgroup analysis based on the cut-off value of CTR, country, source of HR and histology type was conducted to detect the sources of heterogeneity. Statistical analyses were performed using STATA version 12.0.

RESULTS

A total of 29 studies published between 2001 and 2022 with 10,347 patients were enrolled. The pooled results demonstrated that elevated CTR was associated with poorer overall survival (HR = 1.88, 95% CI 1.42-2.50, P < 0.01) and disease-free survival (DFS)/recurrence-free survival (RFS)/progression-free survival (PFS) (HR = 1.42, 95% CI 1.27-1.59, P < 0.01) in NSCLC. According to subgroup analysis by the cut-off value of CTR and histology type, both lung adenocarcinoma and NSCLC patients who had a higher CTR showed worse survival. Subgroup analysis stratified by country revealed that CTR was a prognostic factor for OS and DFS/RFS/PFS in Chinese, Japanese, and Turkish patients.

CONCLUSIONS

In NSCLC patients with high CTR, the prognosis was worse than that with low CTR, indicating that CTR may be a prognostic factor.

Evaluation of response to stereotactic body radiation therapy for nonsmall cell lung cancer: PET response criteria in solid tumors versus response evaluation criteria in solid tumors

OBJECTIVE

Recommendations for surveillance after stereotactic body radiation therapy (SBRT) for early-stage nonsmall cell lung cancer (NSCLC) are not well defined. Recently, PET response criteria in solid tumors (PERCIST) have been proposed as a new standardized method to assess radiotherapeutic response both quantitatively and metabolically. The aim of this study was to evaluate therapeutic response following SBRT in early-stage NSCLC patients by comparing PERCIST with the currently widely used RECIST.

MATERIALS AND METHODS

Forty-nine patients with early-stage NSCLC who had been prescribed SBRT were studied. Responses of lesion were evaluated using CT and 18F-FDG PET according to the RECIST and PERCIST methods. PET-CT scans were obtained before SBRT and 3-6 months after SBRT. Associations between overall survival (OS) and clinicopathologic results (histology, tumor location, tumor size, lymphatic invasion, clinical stage, and radiotherapeutic responses in RECIST and PERCIST) were statistically analyzed. The median patient follow-up was 30 months.

RESULTS

Thirteen patients had stage IA, 9 stage IB, 10 stage IIA, and 17 stage IIB biopsy-proven NSCLC. Three-year OS was 79.6%. CT scans indicated three regional recurrences. PET-CT/chest indicated three regional recurrences and distant metastasis. Significant differences were observed in response classification between RECIST and PERCIST (Wilcoxon signed-rank test, P = 0.0041). Univariate analysis showed that clinical stage, RECIST, and PERCIST were significant factors associated with OS, whereas by multivariate analysis PERCIST was the only predictor of OS. SMD, PMD/PMR, and CMR in PERCIST criteria were indicative of a 9.900-fold increase in the risk of OS in early NSCLC patients [risk ratio, 9.900 (95% CI, 1.040-21.591); P = 0.001].

CONCLUSION

RECIST based on the anatomic size reduction rate did not demonstrate the correlation between radiotherapeutic response and prognosis in patients with early-stage NSCLC receiving SBRT. However, PERCIST was shown as the strongest independent predictor of outcomes. PERCIST might be considered more suitable for the evaluation of NSCLC tumor response to SBRT than RECIST.

Comparison of stereotactic body radiotherapy and radiofrequency ablation for early-stage non-small cell lung cancer: a systematic review and meta-analysis

Background

Stereotactic body radiation therapy (SBRT) and radiofrequency ablation (RFA) are recommended for patients with inoperable early-stage non-small cell lung cancer (NSCLC), with both offering promising results. However, it is largely unknown which of these two treatment modalities provides superior benefits for patients. Therefore, this systematic review and meta-analysis compared clinical outcomes and safety between SBRT and RFA in patients with inoperable early-stage NSCLC.

Methods

Eligible studies published between 2001 and 2020 were obtained through a comprehensive search of the PubMed, Medline, Embase, and Cochrane Library databases. Original English-language studies on the treatment of early-stage NSCLC with SBRT or RFA were included. Local control (LC) rates, overall survival (OS) rates, and adverse events were obtained by pooled analyses.

Results

Eighty-seven SBRT studies (12,811 patients) and 18 RFA studies (1,535 patients) met the eligibility criteria. For SBRT, the LC rates (with 95% confidence intervals) at 1, 2, 3, and 5 years were 98% (97-98%), 95% (95-96%), 92% (91-93%), and 92% (91-93%), respectively, which were significantly higher than those for RFA [75% (69-82%), 31% (22-39%), 67% (58-76%), and 41% (30-52%), respectively] (P<0.01). There were no significant differences in short-term OS between SBRT and RFA [1-year OS rate: 87% (86-88%) versus 89% (88-91%), P=0.07; 2-year OS rate: 71% (69-72%) versus 69% (64-74%), P=0.42]. Regarding long-term OS, the 3- and 5-year OS rates for SBRT were 58% (56-59%) and 39% (37-40%), respectively, which were significantly (P<0.01) superior to those for RFA [48% (45-51%) and 21% (19-23%), respectively]. The most common complication of SBRT was radiation pneumonitis (grade ≥2), making up 9.1% of patients treated with SBRT, while pneumothorax was the most common complication of RFA, making up 27.2% of patients treated with RFA.

Discussion

Compared with RFA, SBRT has superior LC and long-term OS rates but similar short-term OS rates. Prospective randomized trials with large sample sizes comparing the efficacy of SBRT and RFA are warranted.

Prognostic value of metabolic signature on 18F-FDG uptake in breast cancer patients after radiotherapy

Radiotherapy (RT) is a major modality of postoperative treatment in breast cancer. The maximal standardized value (SUVmax) is ¹⁸FDG-PET/CT derived parameter that reported to be a valuable prognostic factor in cancer patients. Herein, we aimed to identify a prognostic gene signature associated with glucose uptake for breast cancer patients after RT by leveraging the mRNA expression profiling on public datasets. The glucose uptake signature was constructed using the single sample gene set enrichment analysis (ssGSEA) algorithm and evaluated in GSE21217 where SUVmax value was measured by PET-CT directly. The prognostic value was validated in three post-RT breast cancer cohorts (GSE103744, NKI, and FUSCC databases). The patients were stratified into glucose uptake signature score-high and low groups. Patients with a higher score had worse survival than those with a lower score. Mechanistically, the glucose uptake signature was calculated in each cell type of a single-cell RNA-seq database from five breast cancer patients. Glucose uptake signature score was significantly elevated in the malignant epithelial cells compared with normal ones. The immunosuppression markers including PDCD1, TIGIT, LAG3, and HAVCR2 were significantly upregulated in the T cells bearing a high glucose uptake signature score. Collectively, our results demonstrated the potential prognostic value of a glucose uptake signature in the post-RT breast cancer patients.

Potential role of functional imaging in predicting outcome for patients treated with carbon ion therapy: a review

OBJECTIVE

Carbon ion radiation therapy (CIRT) is an emerging radiation technique with advantageous physical and radiobiologic properties compared to conventional radiotherapy (RT) providing better response in case of radioresistant and hypoxic tumors. Our aim is to critically review if functional imaging techniques could play a role in predicting outcome of CIRT-treated tumors, as already proven for conventional RT.

METHODS

14 studies, concerning Magnetic resonance imaging (MRI) and Positron Emission Tomography (PET), were selected after a comprehensive search on multiple electronic databases from January 2000 to March 2020.

RESULTS

MRI studies (n = 5) focused on diffusion-weighted MRI and, even though quantitative parameters were the same in all studies (apparent diffusion coefficient, ADC), results were not univocal, probably due to different imaging acquisition protocols and tumoral histology. For PET studies (n = 9), different tracers were used such as [¹⁸F]FDG and other uncommon tracers ([¹¹C]MET, [¹⁸F]FLT), with a relevant heterogeneity regarding parameters used for outcome assessment.

CONCLUSION

No conclusion can be drawn on the predictive value of functional imaging in CIRT-treated tumors. A standardization of image acquisition, multi-institutional large trials and external validations are needed in order to establish the prognostic value of functional imaging in CIRT and to guide clinical practice.

ADVANCES IN KNOWLEDGE

Emerging studies focused on functional imaging's role in predicting CIRT outcome. Due to the heterogeneity of images acquisition and studies, results are conflicting and prospective large studies with imaging standardized protocol are needed.

Assessing tumor angiogenesis using dynamic contrast-enhanced integrated magnetic resonance-positron emission tomography in patients with non-small-cell lung cancer

Background

Angiogenesis assessment is important for personalized therapeutic intervention in patients with non-small-cell lung cancer (NSCLC). This study investigated whether radiologic parameters obtained by dynamic contrast-enhanced (DCE)-integrated magnetic resonance-positron emission tomography (MR-PET) could be used to quantitatively assess tumor angiogenesis in NSCLC.

Methods

This prospective cohort study included 75 patients with NSCLC who underwent DCE-integrated MR-PET at diagnosis. The following parameters were analyzed: metabolic tumor volume (MTV), maximum standardized uptake value (SUV_max), reverse reflux rate constant (k_ep), volume transfer constant (K^trans), blood plasma volume fraction (v_p), extracellular extravascular volume fraction (v_e), apparent diffusion coefficient (ADC), and initial area under the time-to-signal intensity curve at 60 s post enhancement (iAUC₆₀). Serum biomarkers of tumor angiogenesis, including vascular endothelial growth factor-A (VEGF-A), angiogenin, and angiopoietin-1, were measured by enzyme-linked immunosorbent assays simultaneously.

Results

Serum VEGF-A (p = 0.002), angiogenin (p = 0.023), and Ang-1 (p <  0.001) concentrations were significantly elevated in NSCLC patients compared with healthy individuals. MR-PET parameters, including MTV, K^trans, and k_ep, showed strong linear correlations (p <  0.001) with serum angiogenesis-related biomarkers. Serum VEGF-A concentrations (p = 0.004), MTV values (p <  0.001), and k_ep values (p = 0.029) were significantly higher in patients with advanced-stage disease (stage III or IV) than in those with early-stage disease (stage I or II). Patients with initial higher values of angiogenesis-related MR-PET parameters, including MTV > 30 cm³ (p = 0.046), K^trans > 200 10^− 3/min (p = 0.069), and k_ep > 900 10^− 3/min (p = 0.048), may have benefited from angiogenesis inhibitor therapy, which thus led to significantly longer overall survival.

Conclusions

The present findings suggest that DCE-integrated MR-PET provides a reliable, non-invasive, quantitative assessment of tumor angiogenesis; can guide the use of angiogenesis inhibitors toward longer survival; and will play an important role in the personalized treatment of NSCLC.

18F-FDG PET/CT quantitative parameters as prognostic factor in localized and inoperable lung cancer

AIM

To assess the utility of ¹⁸F-FDG PET/CT quantitative parameters as prognostic factor in patients diagnosed with localized and inoperable lung cancer treated by stereotactic body radiotherapy (SBRT).

MATERIAL AND METHODS

Fifty patients (42 men) diagnosed in the last 7years with early-stage lung cancer and treated with SBRT alone were assessed by a prospective study. After PET/CT study, metabolic parameters maximum SUV (SUVmax), metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were determined at different thresholds. The prognostic factors for overall survival (OS), cause-specific survival (CS) and disease-free survival (DFS) were analysed by Cox proportional hazards model and the survival analysis by Kaplan-Meier method.

RESULTS

The average follow-up was 39.6months, with 21 recurrences and 24 dead. Univariate analysis determined MTV₃₀ and MTV₄₀ as predictors for OS; MTV₃₀, MTV₄₀, TLG₃₀ and TLG₄₀ for CS, and MTV₂, MTV₃₀, MTV₄₀, TLG₂, TLG₃₀ and TLG₄₀ for DFS. Survival analysis found statistically significant differences for CS and DFS depending on tumor size and for DFS considering the cut-off values of MTV₂ and TLG₂ (threshold SUVmax=2). SUVmax, age and sex were not shown to be significant factors.

CONCLUSION

Pre-treatment quantitative assessment using metabolic parameters MTV₂ and TLG₂ as well as tumor size proved to be prognostic factors in patients diagnosed with localized and inoperable lung cancer treated by SBRT. Results could help to personalize treatment.