No abstract

Among attempts to delay development of resistance to tyrosine kinase inhibitors (TKIs) in patients with advanced non-small cell lung cancer (NSCLC) with activating mutations of epidermal growth factor receptor (EGFR), intercalated therapy has not been properly evaluated. In a phase II trial, 38 patients with EGFR mutated NSCLC in advanced stage were treated with 4 to 6 3-weekly cycles of intercalated schedule with gemcitabine (1250 mg/m2, days 1 and 4), cisplatin (75 mg/m2, day 2) and erlotinib (150 mg, days 5 – 15), followed by continuous erlotinib as maintenance. In addition to standard radiologic evaluation according to RECIST, PET/CT was done prior to treatment and at 6 months, using PERCIST as a method for assessment of response. The primary endpoint was progression-free survival (PFS). In general, tolerance to treatment was good, even among 8 patients with performance status 2–3 and 13 patients with brain metastases; grade 4 toxicity included 2 cases of neutropenia and 4 thrombo-embolic events. Complete response (CR) or partial response (PR) were seen in 15 (39.5%) and 17 (44.7%) cases, respectively. All cases of CR were confirmed also by PET/CT. Median PFS was 23.4 months and median overall survival (OS) was 38.3  months. After a median follow-up of 35 months, 8 patients are still in CR and on maintenance erlotinib. In conclusion, intercalated treatment for treatment-naive patients with EGFR activating mutations leads to excellent response rate and prolonged PFS and survival. Comparison of the intercalated schedule to monotherapy with TKIs in a randomized trial is warranted.

Background In the setting of primary hyperparathyroidism (PHPT), [18F]fluorocholine PET/CT (FCH-PET) has excellent diagnostic performance, with experienced practitioners achieving 97.7% accuracy in localising hyperfunctioning parathyroid tissue (HPTT). Due to the relative triviality of the task for human readers, we explored the performance of deep learning (DL) methods for HPTT detection and localisation on FCH-PET images in the setting of PHPT. Patients and methods We used a dataset of 93 subjects with PHPT imaged using FCH-PET, of which 74 subjects had visible HPTT while 19 controls had no visible HPTT on FCH-PET. A conventional Resnet10 as well as a novel mPETResnet10 DL model were trained and tested to detect (present, not present) and localise (upper left, lower left, upper right or lower right) HPTT. Our mPETResnet10 architecture also contained a region-of-interest masking algorithm that we evaluated qualitatively in order to try to explain the model’s decision process. Results The models detected the presence of HPTT with an accuracy of 83% and determined the quadrant of HPTT with an accuracy of 74%. The DL methods performed statistically worse (p < 0.001) in both tasks compared to human readers, who localise HPTT with the accuracy of 97.7%. The produced region-of-interest mask, while not showing a consistent added value in the qualitative evaluation of model’s decision process, had correctly identified the foreground PET signal. Conclusions Our experiment is the first reported use of DL analysis of FCH-PET in PHPT. We have shown that it is possible to utilize DL methods with FCH-PET to detect and localize HPTT. Given our small dataset of 93 subjects, results are nevertheless promising for further research.