Synopsis

Lung cancer is characterized by high intra-tumoral heterogeneity, which potentially increases the difficulties of obtaining accurate biopsy results. By combining CT guided-biopsy of lung masses with previously acquired PET/MR images, we investigated the pathological statuses and gene mutation of tissues samples from different locations within one tumor. The results demonstrated that PET/MR can increase the accuracy of a biopsy as well as enhance our understanding of the complicated biological behaviors of cancers.

Introduction

CT-guided needle biopsy is often used in the diagnosis of lung cancer. However, obtaining a representative biopsy sample is challenging since the acquired information without metabolic characteristics of the lesion is far from comprehensive. PET/CT with added value in improving the tissue-sampling accuracy over CT-only guided biopsy has thus been used for lung biopsy, but it is still limited in tissue heterogeneity as some metabolically active benign processes like inflammation and infection can also show high ¹⁸F-FDG uptake^[1]. Therefore, there is always a risk of placing a CT-guided or PET/CT-guided needle into an unrepresentive site of the lesion, leading to inadequate tissue sampling and false-negative results.

Since the introduction of Positron emission tomography/magnetic resonance with ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG PET/MR), the potential of this new hybrid imaging technique to assess tumor aggressiveness and address the issue of intratumoral heterogeneity has gradually come into the community’s insight: morphology is represented by T1/T2 images, metabolism is demonstrated by PET, and cellular function is indicated by diffusion weighted imaging (DWI)^[2].

By combining CT-guided biopsy of lung masses and previously acquired PET/MR images, this study aims to investigate the diversity of pathological statuses and gene mutations at different locations within one tumor in order to determine the utility of applying PET/MR in the assessment of lung tumors’ heterogeneity in the future.

Methods

The Institutional Ethics Committee approved this prospective study. 6 untreated patients with findings on CT highly suggestive of lung cancer with lesional diameters greater than 2.5cm were prospectively recruited and imaged with FDG PET/MR (Biograph mMR, Siemens Healthcare, Erlangen, Germany). MR scans included the following sequences: T1WI, T2WI, DWI with multiple b-values, Dixon water and fat images. CT-guided biopsies were scheduled within 3 days after the PET/MR examination.

Two regions of every tumor were sampled: the region with the highest signal on DWI and the most metabolically active region on PET. Pathological results were analyzed for 54 mutation types and their frequencies. Comparisons were made between samples determined to be active on PET and samples determined to be active on MR with the paired-samples t test.

Results

All 6 patients underwent an ¹⁸F-FDG PET/MR examination successfully. 5 cases were diagnosed as lung cancer by pathology; the other case was pneumonia. In most subjects, there was a rough overlap between areas of abnormal signal on both non-FDG and FDG-avidity, but there were significant differences in the location of the peak signal. For instance, genes CDKN2A, ERBB4, and TP53 of 1 patient had higher mutation frequencies in regions suggested to be most active on PET than regions suggested to be most active on DWI as summarized in Figure 1. The patient with pneumonia and one of 5 patient with lung cancer did not have the gene tests while the relevant genes’mutation frequencies of the other 4 patients differed significantly between puncture site guided by FDG-avidity and by non-FDG(DWI and Dixon fat), as demonstrated in Table 3.(P=0.018)

Discussion

PET/MR can facilitate differentiating malignancy from these benign processes, as MR can provide better soft tissue contrast with T1/T2 images as well as functional images like DWI. PET and DWI are both functional modalities correlated with vital tumor proliferation and aggressiveness from different aspects^[3]. The genetic differences between biopsy samples guided by PET results and DWI results, as shown in Tables 1-2, most likely arise from the differences in the way these images are generated.¹⁸F-FDG reflects the glucose metabolism of tumor cells while DWI is derived from measurement of the random movement (Brownian motion) of water molecule protons. And this is also why we got the statistically significant difference in Table 3. Additionally, Dixon water and fat images may indicate valuable functional information, while weighted images are meaningless for localization of the puncture site.

Despite the limited match of PET/MR registered with intraprocedural CT images, PET/MR images can provide the important references for lung cancer biopsy and reduce the occurrence of false-negative results (Figure 2). This may explain anticancer drugs targeting the aberrant gene still work well, even though gene test shows negative^[4].

This study is limited by its small sample size (n=6) and lack of a control group. In addition, PET/MR images were not registered with intraprocedural CT images, which may potentially influence the localization of the lesion. Further research using PET/MR results to plan CT-guided biopsy sites should be done in larger sample sizes to better understand the imaging correlates of intratumoral heterogeneity.

Conclusion

Acknowledgements

My deepest gratitude goes first and foremost to Professor Jiahe Tian, my supervisor, for his constant guidance and assistance paved the way for my paper. And I would like to express my heartfelt appreciation of the help from my teachers and partners. Additionally, big thanks to my patients for their participation in the research.

Lastly, draw a heart shape for the anonymous reviewers, who constributed consideraly to the publication and improvement of my paper.

References

[1] Purandare N C, Kulkarni A V, Kulkarni S S, et al. 18F-FDG PET/CT-directed biopsy: does it offer incremental benefit?[J]. Nuclear Medicine Communications, 2013, 34(3):203-210.

[2] Metz S, Ganter C, Lorenzen S, et al. Multiparametric MR and PET Imaging of Intratumoral Biological Heterogeneity in Patients with Metastatic Lung Cancer Using Voxel-by-Voxel Analysis. PLoS One. 2015;10(7):e0132386.

[3]Sauter AW, Stieltjes B, Weikert T, et al. The Spatial Relationship betweenApparent Diffusion Coefficient and Standardized Uptake Value of18F-Fluorodeoxyglucose Has a Crucial Influence on the Numeric Correlation of Both Parameters in PET/MRI of Lung Tumors. Contrast Media Mol Imaging.2017；2017:8650853.

[4]Saito M, Shiraishi K, Kunitoh H, Takenoshita S, Yokota J, Kohno T. Gene aberrations for precision medicine against lung adenocarcinoma. Cancer Sci. 2016;107(6):713-720.