LungHist700: A dataset of histological images for deep learning in pulmonary pathology

Accurate detection and classification of lung malignancies are crucial for early diagnosis, treatment planning, and patient prognosis. Conventional histopathological analysis is time-consuming, limiting its clinical applicability. To address this, we present a dataset of 691 high-resolution (1200 × 1600 pixels) histopathological lung images, covering adenocarcinomas, squamous cell carcinomas, and normal tissues from 45 patients. These images are subdivided into three differentiation levels for both pathological types: well, moderately, and poorly differentiated, resulting in seven classes for classification. The dataset includes images at 20x and 40x magnification, reflecting real clinical diversity. We evaluated image classification using deep neural network and multiple instance learning approaches. Each method was used to classify images at 20x and 40x magnification into three superclasses. We achieved accuracies between 81% and 92%, depending on the method and resolution, demonstrating the dataset's utility.

Non-small cell lung cancer in China

In China, lung cancer is a primary cancer type with high incidence and mortality. Risk factors for lung cancer include tobacco use, family history, radiation exposure, and the presence of chronic lung diseases. Most early-stage non-small cell lung cancer (NSCLC) patients miss the optimal timing for treatment due to the lack of clinical presentations. Population-based nationwide screening programs are of significant help in increasing the early detection and survival rates of NSCLC in China. The understanding of molecular carcinogenesis and the identification of oncogenic drivers dramatically facilitate the development of targeted therapy for NSCLC, thus prolonging survival in patients with positive drivers. In the exploration of immune escape mechanisms, programmed cell death protein 1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitor monotherapy and PD-1/PD-L1 inhibitor plus chemotherapy have become a standard of care for advanced NSCLC in China. In the Chinese Society of Clinical Oncology's guidelines for NSCLC, maintenance immunotherapy is recommended for locally advanced NSCLC after chemoradiotherapy. Adjuvant immunotherapy and neoadjuvant chemoimmunotherapy will be approved for resectable NSCLC. In this review, we summarized recent advances in NSCLC in China in terms of epidemiology, biology, molecular pathology, pathogenesis, screening, diagnosis, targeted therapy, and immunotherapy.

Less Is More

OBJECTIVES

Neoadjuvant chemotherapy (NAC) confers a survival advantage for muscle-invasive bladder cancer and is now recommended for chemotherapy-eligible patients. NAC may result in absent gross tumor, and current cystectomy gross examination protocols do not specify approach for these cases.

METHODS

We included cystectomies performed from 2010 to 2018, capturing a period pre- and post-NAC recommendations. Gross descriptions were reviewed and slides of patients who received NAC were evaluated for microscopic tumor, number of blocks with tumor, and location of those blocks.

RESULTS

We identified 239 radical cystectomies for bladder cancer (147 NAC, 92 non-NAC). Gross lesions were not identified for 91 cases. NAC cases had more total blocks submitted (mean, 17.5) compared with non-NAC cases (mean, 16.6). More NAC cases had additional blocks submitted (20 cases) compared with non-NAC cases (2), which were more frequently additional random sections. Of 108 NAC cases with residual carcinoma, only 2 (1.9%) were upstaged on additional random sections.

CONCLUSIONS

At our institution, NAC and non-NAC cases are grossed with similar numbers of initial blocks; however, NAC cases are more likely to submit additional sections of gross lesions and random bladder without significant changes in stage. Our data suggest current gross examination protocols are sufficient for NAC cystectomies.

Postoperative computed tomography of insufflated lung specimens obtained by video-assisted thoracic surgery: detection and margin assessment of pulmonary nodules

Objective

To investigate the utility of computed tomography (CT) scans to detect and assess the margin status of pulmonary nodules that were insufflated after being resected by video-assisted thoracic surgery.

Materials and Methods

This was a novel multicenter study conducted at two national referral centers for thoracic diseases. Patients suspected of having lung cancer underwent video-assisted thoracic surgery for the resection of pulmonary nodules, which were submitted to postoperative CT. Measurements from the CT scans were compared with the results of the histopathological analysis.

Results

A total of 37 pulmonary nodules from 37 patients were evaluated. The mean age of the patients was 65 years (range, 36-84 years), and 27 (73%) were female. A CT analysis of insufflated specimens identified all 37 nodules, and 33 of those nodules were found to have tumor-free margins. The histopathological analysis revealed lung cancer in 30 of the nodules, all with tumor-free margins, and benign lesions in the seven remaining nodules.

Conclusion

Postoperative CT of insufflated suspicious lung lesions provides real-time detection of pulmonary nodules and satisfactory assessment of tumor margins. This initial study shows that CT of insufflated lung lesions can be a valuable tool at centers where intraoperative histopathological analysis is unavailable.

Pleural Invasion Depth of Disseminated Nodules in Patients with Stage IVa or Recurrent Thymoma: Assessment, Curative Impact, and Surgical Outcomes

BACKGROUND

Thymoma patients with pleural dissemination are difficult to manage, and their treatment strategy remains undefined. This study aimed to investigate the clinicopathologic features of these patients, focusing on the association between the depth of pleural invasion and prognosis.

METHODS

Between 2003 and 2019, the study identified 120 disseminated lesions in 20 thymoma patients. Seven patients had de novo stage IVa thymoma and 13 were recurrent cases. Extrapleural pneumonectomy was performed for 8 patients and debulking surgery for 12 patients. Invasion depth of pleural tumors was classified into two groups: when the disseminated tumors invaded the pleura beneath the elastic layer, the tumor was diagnosed as Da, and when the disseminated tumors invaded the pleura beyond the elastic layer, the tumor was diagnosed as Db.

RESULTS

Of 120 nodules, 31 (26%), found in eight patients with recurrent malignancies, were classified as Db. The pathologic status of the surgical margin (PSM) was positive in eight patients, seven of whom had Db nodules. The 5-year overall survival (OS) rate was 100% in the Da group and 75% in the Db group (P = 0.02). The 5-year progression-free survival (PFS) rate was 66.7% in the Da group and 25% in the Db group (P = 0.02). Cox univariate analysis showed that PFS was significantly influenced by the depth of invasion (P = 0.04) and PSM (P = 0.03).

CONCLUSION

Depth of pleural invasion may influence survival outcomes for thymoma patients with pleural dissemination. The patients in this study with Da-disseminated nodules had an increased probability of a longer OS and PFS and tended to achieve negative PSM compared with the patients with Db.

Trends in Accuracy and Comprehensiveness of Pathology Reports for Resected NSCLC in a High Mortality Area of the United States

INTRODUCTION

Complete and accurate pathology reports are vital to postoperative prognostication and management. We evaluated the impact of three interventions across a diverse group of hospitals on pathology reports of postresection NSCLC.

METHODS

We evaluated pathology reports for patients who underwent curative-intent surgical resection for NSCLC, at 11 institutions within four contiguous Dartmouth Hospital Referral Regions in Arkansas, Mississippi, and Tennessee from 2004 to 2020, for completeness and accuracy, before and after the following three quality improvement interventions: education (feedback to heighten awareness); synoptic reporting; and a lymph node specimen collection kit. We compared the proportion of pathology reports with the six most important items for postoperative management (specimen type, tumor size, histologic type, pathologic [p] T-category, pN-category, margin status) across the following six patient cohorts: preintervention control, postintervention with four different combinations of interventions, and a contemporaneous nonintervention external control.

RESULTS

In the postintervention era, the odds of reporting all key items were eight times higher than those in the preintervention era (OR = 8.3, 95 % confidence interval [CI]: 6.7-10.2, p < 0.0001). There were sixfold and eightfold increases in the odds of accurate pT- and pN-category reporting in the postintervention era compared with the preintervention era (pT OR = 5.7, 95 % CI: 4.7-6.9; pN OR = 8.0, 95 % CI: 6.5-10.0, both p < 0.0001). Within the intervention groups, the odds of reporting all six key items, accurate pT category, and accurate pN-category were highest in patients who received all three interventions.

CONCLUSIONS

Gaps in the quality of NSCLC pathologic reportage can be identified, quantified, and corrected by rationally designed interventions.

Interstitial lung abnormalities detected incidentally on CT: a Position Paper from the Fleischner Society

The term interstitial lung abnormalities refers to specific CT findings that are potentially compatible with interstitial lung disease in patients without clinical suspicion of the disease. Interstitial lung abnormalities are increasingly recognised as a common feature on CT of the lung in older individuals, occurring in 4-9% of smokers and 2-7% of non-smokers. Identification of interstitial lung abnormalities will increase with implementation of lung cancer screening, along with increased use of CT for other diagnostic purposes. These abnormalities are associated with radiological progression, increased mortality, and the risk of complications from medical interventions, such as chemotherapy and surgery. Management requires distinguishing interstitial lung abnormalities that represent clinically significant interstitial lung disease from those that are subclinical. In particular, it is important to identify the subpleural fibrotic subtype, which is more likely to progress and to be associated with mortality. This multidisciplinary Position Paper by the Fleischner Society addresses important issues regarding interstitial lung abnormalities, including standardisation of the definition and terminology; predisposing risk factors; clinical outcomes; options for initial evaluation, monitoring, and management; the role of quantitative evaluation; and future research needs.

Downregulation of hsa-microRNA-204-5p and identification of its potential regulatory network in non-small cell lung cancer: RT-qPCR, bioinformatic- and meta-analyses

Background

Pulmonary malignant neoplasms have a high worldwide morbidity and mortality, so the study of these malignancies using microRNAs (miRNAs) has attracted great interest and enthusiasm. The aim of this study was to determine the clinical effect of hsa-microRNA-204-5p (miR-204-5p) and its underlying molecular mechanisms in non-small cell lung cancer (NSCLC).

Methods

Expression of miR-204-5p was investigated by real-time quantitative PCR (RT-qPCR). After data mining from public online repositories, several integrative assessment methods, including receiver operating characteristic (ROC) curves, hazard ratios (HR) with 95% confidence intervals (95% CI), and comprehensive meta-analyses, were conducted to explore the expression and clinical utility of miR-204-5p. The potential objects regulated and controlled by miR-204-5p in the course of NSCLC were identified by estimated target prediction and analysis. The regulatory network of miR-204-5p, with its target genes and transcription factors (TFs), was structured from database evidence and literature references.

Results

The expression of miR-204-5p was downregulated in NSCLC, and the downtrend was related to gender, histological type, vascular invasion, tumor size, clinicopathologic grade and lymph node metastasis (P<0.05). MiR-204-5p was useful in prognosis, but was deemed unsuitable at present as an auxiliary diagnostic or prognostic risk factor for NSCLC due to the lack of statistical significance in meta-analyses and absence of large-scale investigations. Gene enrichment and annotation analyses identified miR-204-5p candidate targets that took part in various genetic activities and biological functions. The predicted TFs, like MAX, MYC, and RUNX1, interfered in regulatory networks involving miR-204-5p and its predicted hub genes, though a modulatory loop or axis of the miRNA-TF-gene that was out of range with shortage in database prediction, experimental proof and literature confirmation.

Conclusions

The frequently observed decrease in miR-204-5p was helpful for NSCLC diagnosis. The estimated target genes and TFs contributed to the anti-oncogene effects of miR-204-5p.

IASLC Multidisciplinary Recommendations for Pathologic Assessment of Lung Cancer Resection Specimens After Neoadjuvant Therapy

Currently, there is no established guidance on how to process and evaluate resected lung cancer specimens after neoadjuvant therapy in the setting of clinical trials and clinical practice. There is also a lack of precise definitions on the degree of pathologic response, including major pathologic response or complete pathologic response. For other cancers such as osteosarcoma and colorectal, breast, and esophageal carcinomas, there have been multiple studies investigating pathologic assessment of the effects of neoadjuvant therapy, including some detailed recommendations on how to handle these specimens. A comprehensive mapping approach to gross and histologic processing of osteosarcomas after induction therapy has been used for over 40 years. The purpose of this article is to outline detailed recommendations on how to process lung cancer resection specimens and to define pathologic response, including major pathologic response or complete pathologic response after neoadjuvant therapy. A standardized approach is recommended to assess the percentages of (1) viable tumor, (2) necrosis, and (3) stroma (including inflammation and fibrosis) with a total adding up to 100%. This is recommended for all systemic therapies, including chemotherapy, chemoradiation, molecular-targeted therapy, immunotherapy, or any future novel therapies yet to be discovered, whether administered alone or in combination. Specific issues may differ for certain therapies such as immunotherapy, but the grossing process should be similar, and the histologic evaluation should contain these basic elements. Standard pathologic response assessment should allow for comparisons between different therapies and correlations with disease-free survival and overall survival in ongoing and future trials. The International Association for the Study of Lung Cancer has an effort to collect such data from existing and future clinical trials. These recommendations are intended as guidance for clinical trials, although it is hoped they can be viewed as suggestion for good clinical practice outside of clinical trials, to improve consistency of pathologic assessment of treatment response.

[Development and deployment of a standardized pathology report in lung cancer, basing on a data management software: The PELICAN software]

INTRODUCTION

PELICAN (« Partager Éfficacement en Laboratoire les Informations des Comptes rendus ANatomopathologiques ») is a software which generates standardized reports and, in parallel, allows to automatically create a database that can be used for research purpose. This application has been used in our laboratory since 2014 for central nervous system tumors. The aim of this work was to extend it to another type of tumor, lung cancer.

MATERIALS AND METHODS

The content of the pathology reports was previously defined using various standards (Société Française de Pathologie, Institut National du Cancer, WHO Classification 2015, …). A double codification was used with SNOMED and ADICAP codes. The PELICAN application is a Microsoft Excel file containing a software specifically developed for pathology laboratories, written in Visual Basic for Applications and respecting the CDA-R2 standard.

RESULTS

After definition of the software specifications, a beta-version was installed in February 2018. After various updates, the 3.19 version was installed in July 2018. Almost all lung cancer surgical pathology reports are now generated with the PELICAN software; a total of 56 reports were validated at the time of writing this manuscript. The medical time for the generation of the report was globally the same or decreased for some pathologists. The secretarial time was greatly reduced.

CONCLUSION

The PELICAN software is an easy to use tool that allows to generate standardized reports in pulmonary pathology and to feed a database that can be easily used for statistical purposes.

Pathologic Assessment After Neoadjuvant Chemotherapy for NSCLC: Importance and Implications of Distinguishing Adenocarcinoma From Squamous Cell Carcinoma

INTRODUCTION

Major pathologic response after neoadjuvant chemotherapy (NAC) for NSCLC has been defined as 10% or less residual viable tumor without distinguishing between histologic types. We sought to investigate whether the optimal cutoff percentage of residual viable tumor for predicting survival differs between lung adenocarcinoma (ADC) and squamous cell carcinoma (SCC).

METHODS

Tumor slides from 272 patients treated with NAC and surgery for clinical stage II-III NSCLC (ADC, n = 192; SCC, n = 80) were reviewed. The optimal cutoff percentage of viable tumor for predicting lung cancer-specific cumulative incidence of death (LC-CID) was determined using maximally selected rank statistics. LC-CID was analyzed using a competing-risks approach. Overall survival was evaluated using Kaplan-Meier methods and Cox proportional hazard analysis.

RESULTS

Patients with SCC had a better response to NAC (median percentage of viable tumor: SCC versus ADC, 40% versus 60%; p = 0.027). Major pathologic response (≤10% viable tumor) was observed in 26% of SCC cases versus 12% of ADC cases (p = 0.004). The optimal cutoff percentage of viable tumor for LC-CID was 10% for SCC and 65% for ADC. On multivariable analysis, viable tumor 10% or less was an independent factor for better LC-CID (p = 0.035) in patients with SCC; in patients with ADC, viable tumor 65% or less was a factor for better LC-CID (p = 0.033) and overall survival (p = 0.050).

CONCLUSIONS

In response to NAC, the optimal cutoff percentage of viable tumor for predicting survival differs between ADC and SCC. Our findings have implications for the pathologic assessment of resected specimens, especially in upcoming clinical trials design.

Spread through Air Spaces (STAS) Is an Independent Predictor of Recurrence and Lung Cancer-Specific Death in Squamous Cell Carcinoma

INTRODUCTION

Spread through air spaces (STAS) is a recently recognized pattern of invasion in lung adenocarcinoma; however, it has not yet been characterized in squamous cell carcinoma (SCC).

METHODS

We reviewed 445 resected stage I to III lung SCCs and investigated the clinical significance of STAS. Cumulative incidence of recurrence and lung cancer-specific death were evaluated by competing risks analyses and overall survival by Cox models.

RESULTS

Of the total 445 patients, 336 (76%) were older than 65 years. Among the 273 patients who died, 91 (33%) died of lung cancer whereas the remaining ones died of competing events or unknown cause. STAS was observed in 132 patients (30%) and the frequency increased with stage. The cumulative incidences of any, distant, and locoregional recurrence as well as lung cancer-specific death were significantly higher in patients with STAS compared with in those without STAS, whereas there was no statistically significant difference in overall survival. In multivariable models for any recurrence and lung cancer-specific death, STAS was an independent predictor for both outcomes (p = 0.034 and 0.016, respectively).

CONCLUSION

STAS was present in one-third of resected lung SCCs. In competing risks analysis in a cohort in which three-fourths of the patients were elderly, STAS was associated with lung cancer-specific outcomes. Our findings suggest that STAS is one of the most prognostically significant histologic findings in lung SCC.

MiR-133a is downregulated in non-small cell lung cancer: a study of clinical significance

Background

Despite present studies which suggested miR-133a as a promising biomarker for several cancers, there still exist no articles concerning the validated clinical significance of miR-133a in non-small cell lung cancer (NSCLC). Therefore, in this study, we targeted the correlation between miR-133a expression and clinicopathological significance in NSCLC patients.

Methods

The expression of miR-133a in 125 cases of NSCLC and their paired adjacent non-cancerous tissues was evaluated by quantitative reverse transcription polymerase chain reaction (qRT-PCR). Meanwhile, the relationship between miR-133a expression and several clinicopathological parameters and patient survival was analyzed.

Results

The relative level of miR-133a was 2.0108 ± 1.3334 in NSCLC tissues, significantly lower than that of the adjacent non-cancerous lung tissues (3.6430 ± 2.2625, P = 0.019). The area under curve (AUC) of low expression of miR-133a to diagnose NSCLC was 0.760 (95% CI: 0.702 ~ 0.819, P < 0.001). MiR-133a expression was negatively correlated to lymphatic metastasis (r = −0.182, P = 0.042), tumor size (r = −0.253, P = 0.04), clinical TNM stages (r = −0.154, P = 0.087), and EGFR protein expression (r = −0.612, P < 0.001).

Conclusions

MiR-133a serves as a tumor-suppressive miRNA in human NSCLC, and its downregulation suggests deterioration in NSCLC patients.

Peripheral lung adenocarcinomas with KRAS mutations are more likely to invade visceral pleura

CONTEXT

Kirsten-RAS (KRAS) mutations play an important role in the carcinogenesis of a subset of lung adenocarcinomas and are associated with poorer prognosis.

OBJECTIVE

To investigate the relationship of KRAS mutation status to the histologic subtype of adenocarcinoma according to the recent classification, patient demographics, tumor size, predominant histologic subtype, nodal status, and visceral pleural invasion, in an attempt to uncover the reason for the worse prognosis associated with KRAS mutation.

DESIGN

A total of 187 consecutive resected lung adenocarcinomas from our institution from 2008 to 2011 that were diagnosed according to the new International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society classification and screened for KRAS mutations were included in the study.

RESULTS

A total of 32% of the adenocarcinomas harbored the KRAS mutation. The median age in the KRAS mutation group was 69 years (range, 43-86 years), and male to female ratio was 1:2.3. The proportion of heavy smokers was significantly higher in tumors with KRAS mutation compared with wild type (83% versus 62%; P = .01). A total of 27% of tumors with KRAS mutation had pleural invasion versus 11% of tumors without KRAS mutation (P = .009). A total of 59 tumor samples were positive for KRAS mutation (25 for G12C, 14 for G12A, 8 for G12V, 7 for G12D, 3 for G12S, and 1 for G12T), and only 3 tumors harbored codon 13 mutations (G13C). Two tumors had double mutations.

CONCLUSIONS

KRAS mutations are more common in heavy smokers, and lung adenocarcinomas with KRAS mutation are more likely to invade the visceral pleura. Increased frequency of visceral pleural invasion may explain in part the worse prognosis associated with KRAS mutations.

Napsin-A, TTF-1, EGFR, and ALK Status Determination in Lung Primary and Metastatic Mucin-Producing Adenocarcinomas

Pulmonary mucin-producing adenocarcinomas may be indistinguishable on conventional histology from a metastasis, as thyroid transcription factor-1 (TTF-1) expression often is lacking and KRAS mutations are widely present even in extrapulmonary sites. Few data have been reported on the diagnostic role of napsin-A and epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) gene alterations in this challenging differential diagnosis. Seventy-seven surgically resected cases, including 53 primary and 24 metastatic tumors from different sites, were evaluated for napsin-A, TTF-1, and ALK by immunohistochemistry and for EGFR mutations by direct sequencing. Overall, napsin-A expression in primary lung mucin-producing adenocarcinomas was 36% (8% mucinous, 17% colloid, 87.5% solid, and 100% signet ring cell) and TTF-1 expression reached an overall figure of 42% (12.5% mucinous, 33% colloid, 87.5% solid, and 100% signet ring cell). Metastatic mucinous adenocarcinomas did not react with napsin-A or with TTF-1. All primary and metastatic tumors lacked EGFR mutations, while a single case of signet ring cell lung adenocarcinoma showed ALK expression and rearrangement at fluorescent in situ hybridization analysis. Napsin-A has a lower sensitivity compared with TTF-1 in primary mucin-producing adenocarcinomas of the lung. However, both antibodies have an absolute specificity, being always negative in metastatic mucinous adenocarcinomas. EGFR mutations and ALK translocation or expression are exceedingly rare in mucin-producing adenocarcinomas of the lung, resulting unnecessary as diagnostic tool in this setting.

Lung Carcinoma Staging Problems

The tumor-node-metastasis (TNM) system is the most commonly used staging system for cancers, including lung cancer. The TNM descriptors and the stage groupings reflect differences in patient prognosis and choices for specific therapies. Generally, the higher the T, N, or M, and the higher the stage grouping, the worse the prognosis is for patients in that category. TNM stage is traditionally the most important factor predicting survival of lung cancer patients.

Pulmonary Neuroendocrine Tumors

Pulmonary neuroendocrine tumors are frequently encountered as small biopsy and resection specimens. Although the World Health Organization (WHO) classification is used as a diagnostic framework globally and in daily practice, it is not without controversy and practical difficulties. The WHO criteria for pulmonary neuroendocrine tumors are reviewed with the recognition that the definitions are based on resected specimens, which are often not the usual clinical circumstances. A practical approach is emphasized, and the differential diagnosis is discussed with particular attention to the context in which diagnostic difficulties are most frequently encountered, such as small biopsies and frozen sections.