br This study has several limitations First FDG uptake
This study has several limitations. First, FDG uptake in a tumor le-sion by PET/CT can be biased since small lesions with low activity might be excluded. The inflammatory lesions or non-neoplastic condi-tions could also be included by chance, which might lead to
Fig. 4. Therapeutic strategy for advanced stage non-small cell lung cancer ac-cording to metabolic tumor burden and tumor biologic aggressiveness/beha-vior.
Black arrows designate positive results, whereas white arrows designate nega-tive results.
MTV, metabolic tumor volume; TLG, total lesion glycolysis; cfDNA, cell free DNA.
overestimation of the tumor volume. Therefore, we carefully defined cut-oﬀ values for neoplastic conditions during MTV/TLG calculation by using patients’ radiologic data (other than PET/CT) and patients’ clin-icopathologic information. In addition, we quantified the MTV and TLG by visually estimating the hypermetabolic foci and confirmed the re-sults with two independent specialists who LY 294002 were blinded to the cfDNA data.
Secondly, serum was used to isolate cfDNA rather than plasma. Compared to plasma, cfDNA isolated from serum tends to be at a higher
concentration because of the extra release of DNA during the coagu-lation process . cfDNA from serum has superior correlation and concordance rates according to various quantification methods . In addition, we analyzed total cfDNA concentration rather than in-vestigating allelic frequencies of NSCLC oncogenic drivers. However, to reduce the bias introduced by interfering factors, we used stringent purification methods to minimize the contamination from serum sam-ples and to obtain high quality cfDNA. Thirdly, only Korean patients were analyzed with the retrospective study design, and hence, ethnic diﬀerences might alter the results. Larger prospective trials including various races are needed for further validation.
In conclusion, cfDNA levels might serve as a prognostic factor pons reflects the biological behavior/aggressiveness of the tumor in ad-vanced NSCLC patients. Since patients with low metabolic tumor burden had a diﬀerent prognosis according to cfDNA levels, monitoring their baseline cfDNA levels and metabolic tumor burden might help to identify patients who need aggressive treatment strategies.
Conception/design: Myung Han Hyun, Eun Seong Lee, Jae Seon Eo, Yeul Hong Kim.
Provision of study materials or patients: Myung Han Hyun, Eun Seong Lee, Jae Seon Eo, Yeul Hong Kim.
Providing study materials or patients: Eun Joo Kang, Yoon Ji Choi, Kyong Hwa Park, Sang Won Shin, Sung Yong Lee, Yeul Hong Kim.
Collection and/or assembly of data: Jae Seon Eo, Sungeun Kim, Eun Joo Kang, Jae Sook Sung, Yoon Ji Choi, Kyong Hwa Park, Sang Won Shin, Sung Yong Lee.
Data analysis and interpretation: Myung Han Hyun, Eun Seong Lee, Jae Seon Eo, Yeul Hong Kim.
Manuscript writing: Myung Han Hyun, Eun Seong Lee, Jae Seon Eo, Eun Joo Kang, Yoon Ji Choi, Kyong Hwa Park, Sang Won Shin, Seong Yong Lee, Yeul Hong Kim. Final approval of manuscript: Myung Han Hyun, Eun Seong Lee, Jae Seon Eo, Sungeun Kim, Eun Joo Kang, Jae Sook Sung, Yoon Ji Choi, Kyong Hwa Park, Sang Won Shin, Sung Yong Lee, Yeul Hong Kim.
Source of funding and support
This research was supported in part by the Korea Health Technology R&D Project of the Korea Health Industry Development Institute, Republic of Korea (HI14C0066; to Y. H. Kim); the Bio & Medical Technology Development Program of the National Research Foundation, Republic of Korea (NRF-2015M3A9D7031070; to Y. H. Kim); and Korea University Grant (K1813121; to J. S. Eo).
Appendix A. Supplementary data