Preoperative CT-based radiomics combined with intraoperative frozen section is predictive of invasive adenocarcinoma in pulmonary nodules: a multicenter study

Objectives

Develop a CT-based radiomics model and combine it with frozen section (FS) and clinical data to distinguish invasive adenocarcinomas (IA) from preinvasive lesions/minimally invasive adenocarcinomas (PM).

Methods

This multicenter study cohort of 623 lung adenocarcinomas was split into training (n = 331), testing (n = 143), and external validation dataset (n = 149). Random forest models were built using selected radiomics features, results from FS, lesion volume, clinical and semantic features, and combinations thereof. The area under the receiver operator characteristic curves (AUC) was used to evaluate model performances. The diagnosis accuracy, calibration, and decision curves of models were tested.

Results

The radiomics-based model shows good predictive performance and diagnostic accuracy for distinguishing IA from PM, with AUCs of 0.89, 0.89, and 0.88, in the training, testing, and validation datasets, respectively, and with corresponding accuracies of 0.82, 0.79, and 0.85. Adding lesion volume and FS significantly increases the performance of the model with AUCs of 0.96, 0.97, and 0.96, and with accuracies of 0.91, 0.94, and 0.93 in the three datasets. There is no significant difference in AUC between the FS model enriched with radiomics and volume against an FS model enriched with volume alone, while the former has higher accuracy. The model combining all available information shows minor non-significant improvements in AUC and accuracy compared with an FS model enriched with radiomics and volume.

Conclusions

Radiomics signatures are potential biomarkers for the risk of IA, especially in combination with FS, and could help guide surgical strategy for pulmonary nodules patients.

Key Points

• A CT-based radiomics model may be a valuable tool for preoperative prediction of invasive adenocarcinoma for patients with pulmonary nodules.

• Radiomics combined with frozen sections could help in guiding surgery strategy for patients with pulmonary nodules.

Electronic supplementary material

The online version of this article (10.1007/s00330-019-06597-8) contains supplementary material, which is available to authorized users.

Marginal radiomics features as imaging biomarkers for pathological invasion in lung adenocarcinoma

OBJECTIVES

Lung adenocarcinomas which manifest as ground-glass nodules (GGNs) have different degrees of pathological invasion and differentiating among them is critical for treatment. Our goal was to evaluate the addition of marginal features to a baseline radiomics model on computed tomography (CT) images to predict the degree of pathologic invasiveness.

METHODS

We identified 236 patients from two cohorts (training, n = 189; validation, n = 47) who underwent surgery for GGNs. All GGNs were pathologically confirmed as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IA). The regions of interest were semi-automatically annotated and 40 radiomics features were computed. We selected features using L1-norm regularization to build the baseline radiomics model. Additional marginal features were developed using the cumulative distribution function (CDF) of intratumoral intensities. An improved model was built combining the baseline model with CDF features. Three classifiers were tested for both models.

RESULTS

The baseline radiomics model included five features and resulted in an average area under the curve (AUC) of 0.8419 (training) and 0.9142 (validation) for the three classifiers. The second model, with the additional marginal features, resulted in AUCs of 0.8560 (training) and 0.9581 (validation). All three classifiers performed better with the added features. The support vector machine showed the most performance improvement (AUC improvement = 0.0790) and the best performance was achieved by the logistic classifier (validation AUC = 0.9825).

CONCLUSION

Our novel marginal features, when combined with a baseline radiomics model, can help differentiate IA from AIS and MIA on preoperative CT scans.

KEY POINTS

• Our novel marginal features could improve the existing radiomics model to predict the degree of pathologic invasiveness in lung adenocarcinoma.

Diagnostic value of contrast-enhanced CT scans in identifying lung adenocarcinomas manifesting as GGNs (ground glass nodules)

This study aimed to explore the value of contrast-enhanced computed tomography (CT) scans in the differential diagnosis of atypical adenomatous hyperplasia, adenocarcinoma in situ, minimal invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC), which manifested as ground glass nodules (GGNs) or mixed GGNs.Unenhanced and contrast-enhanced presurgical CT images of 136 cases of GGNs were compared. The nodules were diagnosed based on their solid portions and maximum dimensions, and the findings obtained using both contrast-enhanced and unenhanced CT images were analyzed in corroboration with the pathological diagnosis.Most (53/56) preinvasive nodules showed increased mean CT values after contrast administration. In the MIA group, after contrast administration, enlargements of the solid portions were seen in 48 nodules (48/60), and the elevation of the mean CT value was observed in 12 nodules (12/60). The vast majority of IAC nodules (29/30) showed enlargement of the solid portions after contrast administration. Moreover, for group differences, findings obtained using this approach statistically match the pathological findings.Contrast-enhanced CT scans are more useful than unenhanced CT scans for the diagnosis of GGNs, especially for pure GGNs, before surgery.

CT-texture analysis of subsolid nodules for differentiating invasive from in-situ and minimally invasive lung adenocarcinoma subtypes

PURPOSE

The purpose of this study was to evaluate the usefulness of computed tomography-texture analysis (CTTA) in differentiating between in-situ and minimally-invasive from invasive adenocarcinomas in subsolid lung nodules (SSLNs).

MATERIAL AND METHODS

Two radiologists retrospectively reviewed 49 SSLNs in 44 patients. There were 27 men and 17 women with a mean age of 63±7 (SD) years (range: 47-78years). For each SSLN, type (pure ground-glass or part-solid) was assessed by consensus and CTTA was conducted independently by each observer using a filtration-histogram technique. Different filters were used before histogram quantification: no filtration, fine, medium and coarse, followed by histogram quantification using mean intensity, standard deviation (SD), entropy, mean positive pixels (MPP), skewness and kurtosis.

RESULTS

We analyzed 13 pure ground-glass and 36 part-solid nodules corresponding to 16 adenocarcinomas in-situ (AIS), 5 minimally invasive adenocarcinomas (MIA) and 28 invasive adenocarcinomas (IVA). At uni- and multivariate analysis CTTA allowed discriminating between IVAs and AIS/MIA (P<0.05 and P=0.025, respectively) with the following histogram parameters: skewness using fine textures and kurtosis using coarse filtration for pure ground-glass nodules, and SD without filtration for part-solid nodules.

CONCLUSION

CTTA has the potential to differentiate AIS and MIA from IVA among SSLNs. However, our results require further validation on a larger cohort.

High versus low attenuation thresholds to determine the solid component of ground-glass opacity nodules

Objectives

To evaluate and compare the diagnostic accuracy of high versus low attenuation thresholds for determining the solid component of ground-glass opacity nodules (GGNs) for the differential diagnosis of adenocarcinoma in situ (AIS) from minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IA).

Methods

Eighty-six pathologically confirmed GGNs < 3 cm observed in 86 patients (27 male, 59 female; mean age, 59.3 ± 11.0 years) between January 2013 and December 2015 were retrospectively included. The solid component of each GGN was defined using two different attenuation thresholds: high (-160 Hounsfield units [HU]) and low (-400 HU). According to the presence or absence of solid portions, each GGN was categorized as a pure GGN or part-solid GGN. Solid components were regarded as indicators of invasive foci, suggesting MIA or IA.

Results

Among the 86 GGNs, there were 57 cases of IA, 19 of MIA, and 10 of AIS. Using the high attenuation threshold, 44 were categorized as pure GGNs and 42 as part-solid GGNs. Using the low attenuation threshold, 13 were categorized as pure GGNs and 73 as part-solid GGNs. The sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy for the invasive focus were 55.2%, 100%, 100%, 22.7%, and 60.4%, respectively, for the high attenuation threshold, and 93.4%, 80%, 97.2%, 61.5%, and 91.8%, respectively, for the low attenuation threshold.

Conclusion

The low attenuation threshold was better than the conventional high attenuation threshold for determining the solid components of GGNs, which indicate invasive foci.

Utility of imaging modalities for predicting carcinogenesis in lobular endocervical glandular hyperplasia

Objectives

To investigate the use of imaging methods for predicting carcinogenesis in lobular endocervical glandular hyperplasia (LEGH).

Methods

We retrospectively analyzed preoperative images on transvaginal sonography and magnetic resonance imaging (MRI) in 23 cases with histologically diagnosed LEGH.

Results

Shape of cervical multicystic lesions on MR images could be divided into two types the flower-type with many small cysts surrounded by larger cysts, and the raspberry-type with many tiny, closely aggregated cysts. Six (46%) of 13 cases had raspberry-type lesions that were not detected on transvaginal sonography but were seen on MRI. Adenocarcinoma in situ (AIS) was identified in 4 postmenopausal women with raspberry-type lesions during the follow-up periods. In these cases, cytologic examination by targeted endocervical sampling using sonography enabled early detection of AIS.

Conclusions

MRI and cytologic examination by targeted endocervical sampling may be very useful for predicting carcinogenesis in LEGH.

Adenocarcinoma in situ and minimally invasive adenocarcinoma in lungs of smokers: image feature differences from those in lungs of non-smokers

PURPOSE

We aimed to examine the characteristics of imaging findings of adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA) in the lungs of smokers compared with those of non-smokers.

MATERIALS AND METHODS

We included seven cases of AIS and 20 cases of MIA in lungs of smokers (pack-years ≥ 20) and the same number of cases of AIS and MIA in lungs of non-smokers (pack-years = 0). We compared the diameter of the entire lesion and solid component measured on computed tomography (CT) images, pathological size and invasive component diameter measured from pathological specimens, and CT values of the entire lesion and ground-glass opacity (GGO) portions between the smoker and non-smoker groups.

RESULTS

The diameters of AIS and MIA on CT images and pathological specimens of the smoker group were significantly larger than those of the non-smoker group (p = 0.036 and 0.008, respectively), whereas there was no significant difference in the diameter of the solid component on CT images or invasive component of pathological specimens between the two groups. Additionally, mean CT values of the entire lesion and GGO component of the lesions in the smoker group were significantly lower than those in the non-smoker group (p = 0.036 and 0.040, respectively).

CONCLUSION

AIS and MIA in smoker's lung tended to have larger lesion diameter and lower internal CT values compared with lesions in non-smoker's lung. This study calls an attention on smoking status in CT-based diagnosis for early stage adenocarcinoma.

The value of percentile base on computed tomography histogram in differentiating the invasiveness of adenocarcinoma appearing as pure ground-glass nodules

To investigate the value of percentile base on computed tomography (CT) histogram analysis for distinguishing invasive adenocarcinoma (IA) from adenocarcinoma in situ (AIS) or micro invasive adenocarcinoma (MIA) appearing as pure ground-glass nodules.A total of 42 cases of pure ground-glass nodules that were surgically resected and pathologically confirmed as lung adenocarcinoma between January 2015 and May 2019 were included. Cases were divided into IA and AIS/MIA in the study. The percentile on CT histogram was compared between the 2 groups. Univariate and multivariate logistic regression were used to determine which factors demonstrated a significant effect on invasiveness. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) was used to evaluate the predictive ability of individual characteristics and the combined model.The 4 histogram parameters (25th percentile, 55th percentile, 95th percentile, 97.5th percentile) and the combined model all showed a certain diagnostic value. The combined model demonstrated the best diagnostic performance. The AUC values were as follows: 25th percentile = 0.693, 55th percentile = 0.706, 95th percentile = 0.713, 97.5th percentile = 0.710, and combined model = 0.837 (all P < .05).The percentile of histogram parameters help to improve the ability to radiologically determine the invasiveness of lung adenocarcinoma appearing as pure ground-glass nodules.

The morphological changes of bronchovascular bundles within subsolid nodules on HRCT correlate with the new IASLC classification of adenocarcinoma

AIM

To observe the morphological changes of bronchovascular bundles within subsolid nodules on high-resolution (HR) computed tomography (CT) and analyse the correlation with the new adenocarcinoma classification.

MATERIALS AND METHODS

Two hundred and sixteen lesions (absent consolidation on mediastinal window) were reviewed retrospectively. CT features including dimensions, contour, morphological changes of the blood vessels, and bronchi/bronchioles, vacuole signs, and their correlation with histopathology were evaluated.

RESULTS

Excluding nine non-cancerous lesions, 34 pre-invasive lesions (PILs) including 15 atypical adenomatous hyperplasias (AAHs) and 19 adenocarcinomas in situ (AISs), 21 minimally invasive adenocarcinomas (MIAs), and 152 invasive adenocarcinomas (IACs) were analysed. Lepidic, acinar, and papillary patterns were identified in this cohort of adenocarcinomas. IACs were grouped into three types: type I (lepidic pattern ≥80%, n=47), type II (lepidic pattern ≥50%, <80%, n=67), and type III (lepidic pattern <50%, n=38). The contour of lesions, and morphological changes in vessels and bronchi/bronchioles significantly correlated with the classification of PIL, MIA, and IACs (p=0.000, p=0.000, and p=0.017, respectively). In IACs, the prevalence of vascular abnormalities on HRCT significantly correlated with (p=0.000) the proportion of non-lepidic pattern (23.40% in type I, 58.21% in type II, and 76.32% in type III); the prevalence of bronchial/bronchiolar abnormalities was higher (p=0.008) in type II/III (20.95%) compared with type I (6.38%).

CONCLUSIONS

The morphological changes of vessels and bronchi/bronchioles within the subsolid nodules on HRCT help to differentiate IAC from PIL and MIA, and are more common in non-lepidic predominant adenocarcinomas.

Clinical significance of the lymphoscintigraphy in the evaluation of non-axillary sentinel lymph node localization in breast cancer

BACKGROUND

Identification and biopsy of the sentinel lymph node (SLN) in early-stage breast cancer (T1-T2N0) has become the standard method in the surgical treatment of the axilla, due to its accuracy in the evaluation of axillary lymph node status,thus avoiding extensive axillary lymph node dissection inpatients with negative SLN. The aim of our study is to highlight the role of 99mTc-nanocolloid lymphoscintigraphy in the preoperative lymphatic mapping, especially for SLN localizations outside the axilla, as well as the benefits of this technique in the accurate staging of breast cancer.

MATERIALS AND METHOD

430 patients (age 31-81 years) with breast cancer (T1-T2N0 stage) were included in the study group, who underwent lymphoscintigraphy in order to identify the sentinel lymph node in the Nuclear Medicine Department of "€œProf. Dr. Al. Trestioreanu"€ Institute of Oncology, Bucharest,between October 2008 - July 2014. Radiocolloid (99mTc-nanocolloid)was injected using peritumoral or periareolar intradermal technique, doses between 20-37 MBq (0.3-0.5 ml volume), followed by static and dynamic post-injection acquisitions.Intraoperative identification of the SLN was performed using a gamma-probe, guided by the skin marker performed preoperatively after completion of lymphoscintigraphy.

RESULTS

697 sentinel lymph nodes were identified through imaging techniques in 427 patients (99%). Of them, 364 patients had axillary localization of the SLN, while 48 patients(11%) had non-axillary (pectoral, internal mammary, supraclavicular,intra-mammary) localization and 15 patients (3%)had multiple localization (axillary and non-axillary).Intraoperative histopathological exam revealed lymphatic invasion in 74 SLN (12% macrometastases and 88% micrometastases).

CONCLUSIONS

The identification and biopsy of the sentinel lymph node in stages I and IIA is a useful routine for accurate breast cancer staging, suited for axillary lymphatic drainage, as well as for unusual non-axillary SLN localization, guiding the clinician for further postoperative management of these patients.

Pulmonary Alveolar Proteinosis-Like Pathological Changes Mimicking Lung Adenocarcinoma in Situ

An enlarging ground-glass nodule (GGN) in the lungs closely resembles the characteristic appearance of a well differentiated lung adenocarcinoma or adenocarcinoma in situ (AIS). Herein, we present an unusual case characterized by clinical features suggestive of AIS but pathologically confirmed as exhibiting pulmonary alveolar proteinosis (PAP)-like changes. Patients with enlarging pure GGNs warrant consideration for diagnostic and curative surgery. While a considerable proportion of such cases receives a pathological diagnosis of lung malignancy, it is imperative to consider alternative benign conditions in the differential diagnosis, such as PAP-like changes.

A Case With Gallbladder Villous Tubular Adenoma With High-Grade Intraepithelial Neoplasia Combined With Rectal Villous Tubular Adenocarcinoma

We present the rare case of a 79-year-old female patient with the coexistence of gallbladder villous tubular adenoma and rectal villous tubular adenocarcinoma. She initially presented with blood in the stool, prompting a diagnosis of rectal cancer. Further examination revealed findings suggestive of a gallbladder pathology. Subsequent imaging and histopathological analyses confirmed the presence of a gallbladder villous tubular adenoma, high-grade intraepithelial neoplasia, and rectal villous tubular adenocarcinoma. Surgical interventions included laparoscopic radical resection of the rectal cancer and cholecystectomy. Immunohistochemical examination differentiated between the two distinct adenomatous conditions. This case underscores the importance of thorough evaluation and multidisciplinary approaches for accurate diagnosis and management of complex presentations of gastrointestinal neoplasms.

Clinical and Sonographic Features of Noninvasive Follicular Thyroid Neoplasm With Papillary-Like Nuclear Features: A Retrospective Study

ABSTRACT

This study was designed to investigate the clinical and sonographic features of noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTPs) as compared with classical papillary thyroid carcinoma (cPTC), follicular adenoma (FA), and follicular thyroid carcinoma (FTC). A total of 178 patients were enrolled in this study. The clinical characteristics and sonographic features of thyroid nodules were compared between NIFTP and cPTC or FA/FTC. All nodules were reclassified according to the Thyroid Ultrasound Imaging Reporting and Data System and American Thyroid Association guidelines classification. The mean size of NIFTP was 29.91 ± 14.71 mm, which was larger than that of cPTC ( P = 0.000). Significant difference was found in lymph node metastases between NIFTP and cPTC ( P = 0.000). Most NIFTPs showed solid composition, hypoechoic echogenicity, smooth margin, wider than tall shape, none echogenic foci, absence of halo, and perinodular vascularity, which were similar with FA and FTC. Compared with NIFTP, hypoechoic and very hypoechoic, taller than wide, irregular margin, punctate echogenic foci, absence of halo, and low vascularity were more commonly observed in cPTC. There were statistical differences both in American College of Radiology Thyroid Ultrasound Imaging Reporting and Data System and in American Thyroid Association classification between NIFTP and cPTC ( P < 0.05), but there were no significant differences between NIFTP and FTC/FA ( P > 0.05). The ultrasonographic characteristics of NIFTP were obviously different from cPTC but overlapped with FTC and FA. Ultrasound could help increase preoperative attention of NIFTP in an appropriate clinical setting, which may lead to a more conservative treatment approach.

Quantitative analysis of imaging characteristics in lung adenocarcinoma in situ using artificial intelligence

BACKGROUND

With the rising incidence of pulmonary nodules (PNs), lung adenocarcinoma in situ (AIS) is a critical early stage of lung cancer, necessitating accurate diagnosis for early intervention. This study applies artificial intelligence (AI) for quantitative imaging analysis to differentiate AIS from atypical adenomatous hyperplasia (AAH) and minimally invasive adenocarcinoma (MIA), aiming to enhance clinical diagnosis and prevent misdiagnosis.

METHODS

The study analyzed 1215 PNs with confirmed AAH, AIS, and MIA from six centers using the Shukun AI diagnostic module. Parameters evaluated included demographic data and various CT imaging metrics to identify indicators for clinical application, focusing on the mean CT value's predictive value.

RESULTS

Significant differences were found in several parameters between AAH and AIS, with nodule mass showing the highest predictive value. When comparing AIS to MIA, total nodule volume was the best predictor, followed by the maximum CT value.

CONCLUSION

The mean CT value has limited discriminative power for AIS diagnosis. Instead, the maximum CT value and maximum 3D diameter are recommended for clinical differentiation. Nodule mass and volume of solid components are strong indicators for differentiating AIS from AAH and MIA, respectively.

Correlation between imaging features of pure ground-glass opacities and pathological subtypes of lung minimally invasive adenocarcinoma and precursor lesions

This study aimed to investigate the relationship between imaging features of pure ground-glass opacities (pGGOs) and the pathological subtypes of minimally invasive adenocarcinoma (MIA) and precursor lesions. A retrospective analysis was conducted on data from 1521 patients diagnosed with GGOs as lung adenocarcinoma or precursor lesions between January 2015 and March 2021. The pGGOs were categorized into atypical adenomatous hyperplasia (AAH) / adenocarcinoma in situ (AIS) and MIA groups. Clinical information and CT imaging features were collected. Statistical analysis, logistic regression, and receiver operating characteristic (ROC) curve analysis were performed. A total of 127 patients with 139 pGGOs were included. Maximum radiodensity, minimum radiodensity, mean radiodensity, variance, and skewness showed significant differences between the two groups. Maximum radiodensity and maximum cross-sectional area were identified as risk factors for pathology. The logistic regression model yielded an area under the curve (AUC) of 0.747 (95% CI, 0.666-0.816) for predicting pathological subtypes. The intensity features of pGGOs were found to be significantly different between AAH/AIS and MIA groups. Maximum radiodensity and skewness were independent risk factors for pathology. However, these features did not exhibit satisfactory diagnostic efficiency.

An artificial intelligence algorithm for the detection of pulmonary ground-glass nodules on spectral detector CT: performance on virtual monochromatic images

BACKGROUND

This study aims to assess the performance of an established an AI algorithm trained on conventional polychromatic computed tomography (CT) images (CPIs) to detect pulmonary ground-glass nodules (GGNs) on virtual monochromatic images (VMIs), and to screen the optimal virtual monochromatic energy for the clinical evaluation of GGNs.

METHODS

Non-enhanced chest SDCT images of patients with pulmonary GGNs in our clinic from January 2022 to December 2022 were continuously collected: adenocarcinoma in situ (AIS, n = 40); minimally invasive adenocarcinoma (MIA, n = 44) and invasive adenocarcinoma (IAC, n = 46). A commercial CAD system based on deep convolutional neural networks (DL-CAD) was used to process the CPIs, 40, 50, 60, 70, and 80 keV monochromatic images of 130 spectral CT images. AI-based histogram parameters by logistic regression analysis. The diagnostic performance was evaluated by the receiver operating characteristic (ROC) curves, and Delong's test was used to compare the CPIs group with the VMIs group.

RESULTS

When distinguishing IAC from MIA, the diagnostic efficiency of total mass was obtained at 80 keV, which was superior to those of other energy levels (P < 0.05). And Delong's test indicated that the differences between the area-under-the-curve (AUC) values of the CPIs group and the VMIs group were not statistically significant (P > 0.05).

CONCLUSION

The AI algorithm trained on CPIs showed consistent diagnostic performance on VMIs. When pulmonary GGNs are encountered in clinical practice, 80 keV could be the optimal virtual monochromatic energy for the identification of preoperative IAC on a non-enhanced chest CT.