Skip to main content
Fig. 2 | Molecular Cancer

Fig. 2

From: Targeted next-generation sequencing of circulating free DNA enables non-invasive tumor detection in myxoid liposarcomas

Fig. 2

Quantification of ctDNA in patients’ plasma samples. A Quantification of ctDNA in 9 plasma samples of patient 1 collected during 2 years of treatment. ctDNA was determined by NGS (standard panel) and PIK3CA mutations (c.1624G > A and c.3140A > G) were additionally quantified by ddPCR. He initially presented with a localized MLS of the thigh which was completely resected. Soon after, he developed metastatic disease with predominantly osseous lesions. He then received radiotherapy of bone metastasis and several courses of chemotherapy. Repeated imaging during follow-up showed numerous new skeletal lesions and the patient again received radiotherapy to selected metastasis. He succumbed to his disease 1.5 years after removal of the primary tumor. ctDNA increased to 172 copies/ml (sample 6) when metastatic disease was detected and decreased during radio/chemotherapy to 14 copies/ml (sample 7) and 7 copies/ml (sample 8). There was a rapid incline in ctDNA when multiple new metastases were detected (sample 9: 552 copies/ml). Standard imaging, which reflects the total mass of viable and necrotic tumor cells, showed a steady increase (blue area). The irradiated tumor volume is depicted as a surrogate marker for the necrotic tumor mass (green area). t (12;16) ctDNA levels were higher than PIK3CA ctDNA concentrations. This most likely reflect intertumor heterogeneity with only a fraction of metastases carrying PIK3CA mutations (Fig. 1 D and Supplementary Fig. 4). B Additional target mutations from exome sequencing increase sensitivity of ctDNA detection. Tumor 2 was subjected to exome sequencing to identify additional target mutations. Together with breakpoints and mutations from the standard panel, a 7320 bp hybrid exome panel targeting 15 genomic regions was designed. ctDNA in plasma obtained during treatment was determined by the standard and exome panel. He initially received neo-adjuvant radiotherapy to an MLS of his right thigh and subsequently the tumor was completely resected. Two plasma samples were collected prior to commencement of radiotherapy, a third sample before surgery and a fourth sample after tumor resection. ctDNA quantified by the exome panel (red line) was present in similar amounts at the two time points before treatment, declined after radiotherapy and was not detectable after tumor resection. The standard panel (dashed black line) could detect ctDNA only in the first sample, showing reduced sensitivity compared to the exome panel. The blue area represents the tumor volume as calculated from the MRI scans. C Comparison of different assays in detecting MLS tumor-DNA. Dilution series of MLS tumor-DNA from two tumors (patient 2 and 3) in matched normal DNA were analyzed by ddPCR (PIK3CA mutations p.N345K, c.1035 T > A and p.E545G, c.1634A > G), the NGS standard panel and respective exome panels. Depicted are mean values and linear regression of n = 2 tumors for ddPCR, n = 2 for the standard panel and n = 2 for the exome panels. We observed a similar performance for ddPCR and the standard panel, whereas detection of tumor-DNA with exome panels was clearly superior. D Patient 3 presented with two small localized tumors of his legs (red) after numerous prior resections at another hospital. The tumors were completely resected, but he repeatedly developed local recurrences at both locations in the following years. E These recurrences were subsequently resected at four consecutive operations before a small lung metastasis (0.3 cm3) was detected and subsequently removed. Exome panels were obtained from sequencing one of the primary lesions and the lung metastasis. ctDNA was subsequently quantified with both panels in 15 plasma samples obtained during the course of his treatment. During multifocal localized disease, ctDNA values undulated at low concentrations depending on the presence of viable tumor tissue. The exome panel from the primary tumor best reflected the clinical course (enlarged image section). There was one outlier (circle). Despite complete tumor resections, ctDNA values never reached the baseline indicating MRD. In contrast the plasma sample obtained shortly before resection of the lung metastasis showed markedly increased ctDNA with a decline after its resection. The red line represents ctDNA values measured by the exome panel from the primary lesion and the dashed line ctDNA measured by the exome panel obtained from the lung metastasis. The blue area depicts the tumor volume

Back to article page