Correlation exploration among CT imaging, pathology and genotype of pulmonary ground‐glass opacity

Abstract To analyse the clinical features, imaging manifestation, pathological typing and genetic testing results of patients undergoing surgery for ground‐glass opacity (GGO) nodules, and explore the reasonable diagnosis and treatment program for GGO patients as to provide the basis for the establishment of GGO treatment process. This study is an exploratory study. 465 cases with GGO confirmed by HRCT, undergoing surgery and approved by pathologic diagnosis in Shanghai pulmonary hospital were enrolled in this study. All the patients with GGO were cases with single lesion. The relationship between the clinical, imaging, pathological and molecular biological data of single GGO were statistically studied. (1) Among 465 cases, the median age was 58 years and females were 315 (67.7%); there were 397 (85.4%) non‐smoking, and 354 cases (76.1%) had no clinical symptoms. There were 33 cases of benign and 432 cases of malignant GGO. Significant differences were observed on the size, vacuole sign, pleural indentation and blood vessel sign of GGO between two groups (p < 0.05). Of 230 mGGO, there were no AAH, 13 cases of AIS, 25 cases of MIA and 173 cases of invasive adenocarcinoma. The probability of solid nodules in invasive adenocarcinoma was higher than that in micro invasive carcinoma, and the difference was statistically significant (p < 0.05). 360 cases were followed up with the average follow‐up time of 6.05 months, and GGO of 34 cases (9.4%) increased. (2) In 428 adenocarcinoma samples approved by pathologic diagnosis, there were 262 (61.2%) lesions of EGFR mutation, 14 (3.3%) lesions of KRAS mutation, 1 (0.2%) lesion of Braf mutation, 9 (2.1%) lesions of EML4‐ALK gene fusion and 2 (0.5%) lesions of ROS1 fusion. The detection rate of gene mutation in mGGO was higher than that of pGGO. During the follow‐up period, genetic testing results of 32 GGO showed that EGFR mutation rate was 53.1%, ALK positive rate of 6.3%, KRAS mutation rate of 3.1% and no ros1 and BRAF gene mutation. No statistically significant difference was observed in comparison with unchanged GGO. (3) EGFR mutation rate of invasive adenocarcinoma was the highest (168/228, 73.7%), mainly in the 19Del and L858R point mutations. No KRAS mutation was found in atypical adenoma hyperplasia. No significant difference was observed on the mutation rate of KRAS between different types of GGO (p = 0.811). EML4‐ALK fusion gene was mainly detected in invasive adenocarcinoma (7/9). GGO tends to occur in young, non‐smoking women. The size of GGO is related to the degree of malignancy. Pleural depression sign, vacuole sign and vascular cluster sign are all characteristic images of malignant GGO. pGGO and mGGO reflect the pathological development of GGO. During the follow‐up, it is found that GGO increases and solid components appear, which is the indication of surgical resection. The detection rate of EGFR mutations in mGGO and invasive adenocarcinoma is high. pGGO has heterogeneity in imaging, pathology and molecular biology. Heterogeneity research helps to formulate correct individualized diagnosis and treatment plans.


| INTRODUC TI ON
Since the ground-glass opacity (GGO) was discovered by CT in the 1990s, with the widespread application of HRCT in lung cancer screening, the detection of pulmonary GGO has continued to increase in recent years. 1 GGO is defined as an area of indistinct dense opacity on HRCT in which bronchial or pulmonary vascular structures are still visible, with a maximum diameter of less than 3 cm.
Focal pulmonary GGO is thought to be closely associated with lung adenocarcinoma. 2 How to properly handle GGO and make timely diagnosis and reasonable treatment for early stage lung cancer has become one of the hot spots in clinical oncology. The molecular biology of GGO is a research hotspot in recent years, and the detection of lung cancer driver genes has attracted the attention and discussion of many scholars in the field of GGO.
Several trials have shown that non-mucinous tumours, especially in adenocarcinomas with pathological types of papillary, micropapillary and adherent growth characteristics, have a high EGFR mutation rate, and the EGFR mutation rate in AIS/MIA/LPA can even be as high as 45%. [6][7][8] However, the mechanism of EGFR mutation in GGO lesions is still unclear. Some studies have confirmed that adherent tumours with GGO on imaging are prone to EGFR mutation, and the EGFR mutation rate tends to be related to the volume percentage of GGO, that is, the proportion of GGO in the lesion. The larger the ratio, the higher the EGFR mutation rate. 9,10 However, some studies have shown that EGFR mutations have nothing to do with GGO components. [11][12][13] More research is needed to confirm this. Based on the above studies, the relationship between GGO imaging, pathology, and molecular biology is the key to establishing a GGO diagnosis and treatment strategy, which requires in-depth research.
The detection rate of GGO is increasing year by year with the advancement of imaging technology and the popularisation of lung cancer screening. GGO can be a similar manifestation of many different diseases, but the prognosis of benign and malignant GGO is different. Early and reasonable treatment of malignant lesions can significantly improve the prognosis of patients. Large-scale studies have shown that patients with AIS and MIA have a very low rate of lymph node metastasis, and if they receive radical surgery, their disease-free survival rate can be 100% or close to 100%. 14 Combined with HRCT imaging features, the application of new pathological classification and molecular biology detection are helpful for differential diagnosis and individualized treatment of GGO. This research topic takes GGO patients as the research object, and comprehensively evaluates through imaging, pathology, molecular biology and other means, jointly detects and analyses EGFR, KRAS, ALK fusion genes, ROS1 and BRAF genes, and analyses and studies the heterogeneity of multiple GGOs, to provide a reasonable solution for the follow-up and diagnosis and treatment of GGO, and ultimately serve the clinic and benefit the patients. adenocarcinoma (7/9). GGO tends to occur in young, non-smoking women. The size of GGO is related to the degree of malignancy. Pleural depression sign, vacuole sign and vascular cluster sign are all characteristic images of malignant GGO. pGGO and mGGO reflect the pathological development of GGO. During the follow-up, it is found that GGO increases and solid components appear, which is the indication of surgical resection. The detection rate of EGFR mutations in mGGO and invasive adenocarcinoma is high. pGGO has heterogeneity in imaging, pathology and molecular biology. Heterogeneity research helps to formulate correct individualized diagnosis and treatment plans.

K E Y W O R D S
ground-glass opacity (GGO), lung/pathology, oncogenes, tomography, X-ray computed The purpose of this study is to analyse the clinical features, imaging manifestation, pathological typing and genetic testing results of patients undergoing surgery for ground-glass opacity (GGO) nodules, and explore the reasonable diagnosis and treatment program for GGO patients as to provide the basis for the establishment of GGO treatment process.  Table 1.      The specific pathological morphology of each subtype is shown in   Table 6. The detailed results are shown in Table 7 Table 8. Both ROS1-positive GGOs and 1 BRAF-positive GGO did not increase the maximum diameter of GGOs during follow-up.  Table 9.

| The relationship between GGO pathological types and molecular biological features
Further statistical analysis of different subtypes in IA showed that the invasive adenocarcinoma with adherent growth, acinar-like growth and papillary-like growth had a higher EGFR mutation rate, among which 19Del and L858R point mutations were the main ones.
The results are shown in Table 10.
No KRAS mutation was found in atypical adenomatous hyperplasia, and there was no significant difference between GGO adenocarcinoma types and KRAS mutation rate. The results are shown in

| The relationship between pathological types and ROS1 gene mutation
There were 2 cases of ROS1-positive GGO, 1 case of adenocarcinoma in situ and 1 case of invasive adenocarcinoma with acinar-like growth type.

| The relationship between pathological types and BRAF gene mutation
One case of BRAF-positive GGO was an invasive adenocarcinoma with acinar-like growth.

| DISCUSS ION
The detection rate of GGO is increasing year by year, and it is more likely to occur in young, female, and non-smoking patients. Patients with focal GGO are usually not accompanied by special clinical symptoms at the time of discovery. 15, 16 We use HRCT scanning to make focal GGO in the lung show more detailed CT signs, which is very helpful for qualitative diagnosis. According to the correlation study between CT images and pathological results, the maximum diameter of GGO is related to the pathological type. Signs such as 'pleural indentation sign', 'vascular sign' and 'vacuole sign' are more common in malignant GGO, but in practical work, Not only one of them can be used as the basis for diagnosis, but a comprehensive analysis should be carried out. pGGO and mGGO reflect the pathological development of lung adenocarcinoma GGO, which can help us in preoperative imaging diagnosis and follow-up planning. During follow-up, GGO was found to be enlarged, which is an indication for surgical resection.
EGFR, ALK, BRAF, KRAS, ROS1 are the five common genes in lung adenocarcinoma, and EGFR gene mutation is the most common. The dominant population is female and non-smoking. 17 Among the imaging features, the mGGO mutation rate was high, and the L858R point mutation was the main one, followed by the exon 19 deletion mutation. 18 Multivariate analysis showed that the size of GGO was positively correlated with the mutation rate. It was found that KRAS mutation (codon 12) and EGFR mutation (L858R missense mutation) coexisted in one GGO lesion. More research is needed to TA B L E 3 Imaging comparison in different adenocarcinoma subgroups.  Although this study adopted a sufficient sample size to verify the conclusions in the study, it did not calculate and prove the sample size when selecting the sample, which is one of the limitations of this study. In addition, although this study has obtained fairly detailed results, more experiments are still needed to carry out more in-depth studies to explore GGO.

| CON CLUS ION
The

Medical Innovation Research Project of Shanghai Science and
Technology Commission (22Y11901500).

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declared that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data was available on request from the corresponding author.