Parallel assessment of hypoxia in tumor and LN metastases increases prognostic value of hypoxia-specific PET imaging in locally advanced head-and-neck cancer - secondary analysis of the DDFMISO-trial


Parallel assessment of hypoxia in tumor and LN metastases increases prognostic value of hypoxia-specific PET imaging in locally advanced head-and-neck cancer - secondary analysis of the DDFMISO-trial

Bandurska-Luque, A.; Löck, S.; Haase, R.; Zöphel, K.; Abolmaali, N.; Seidlitz, A.; Perrin, R.; Richter, C.; Troost, E.; Steinbach, J.

Background: Primary tumor (Tu) hypoxia based on hypoxia-specific PET-imaging is a known prognostic parameter for locally-advanced head-and-neck cancer patients. A secondary analysis of the prospective clinical trial on repeated pre- and per-treatment [18F]fluoromisonidazole (FMISO) PET/CT imaging aimed to assess whether parallel evaluation of the oxygenation status in lymph node metastases (LN) and the Tu increases its prognostic value.
Patients and methods: Patients with LN-positive disease from the trial (NCT00180180, Zips et al. 2012, Seidlitz et al. 2015) were included in this analysis (n=45). The patients were treated with curatively intended radiochemotherapy (RCT). The imaging protocol consisted of FMISO PET/CT at four time points: baseline, week 1, 2 and 5. Delineation of the Tu and LNs was based on pre-treatment FDG PET/CT. Qualitative hypoxia analysis was performed for each Tu and LN using a visual binary scale: hypoxic or normoxic being FMISO uptake higher than or equal to background respectively. Based on this scale two prognostic parameters were defined: Tu hypoxia (patients with a hypoxic Tu, independently of LN oxygenation status) and synchronous Tu- and LN-hypoxia (Tu&LN-hypoxia). In the patients with a large LN (n=15) a quantitative analysis of FMISO PET/CT was performed to validate the qualitative hypoxia scale. The log-rank test and multivariate Cox-regression were used to evaluate the prognostic impact of hypoxia on local control (LC) and loco-regional control (LRC).
Results: Qualitative FMISO assessment (Table 1) confirmed poor LC in patients with Tu hypoxia in week 2 and 5. Detection of synchronous Tu- and LN-hypoxia had a strong negative impact on LC and LRC at all measured time-points. These results were supported by multivariate analysis (for LRC: HR=14.8, p=0.016; HR=8.3, p=0,003 and HR=5.5, p=0,005 at baseline, in week 2 and 5, respectively). Moreover, there was a significant correlation between the qualitative and quantitative FMISO PET/CT parameters (p<0.001; R>0.6-0.8).
Conclusions: Parallel evaluation of tumor and LN hypoxia improved the prognostic information in comparison to primary tumor assessment alone, based on secondary analysis of the Dresden FMISO PET/CT trial. If this prognostic value of synchronous tumor- and LN-hypoxia is confirmed in ongoing prospective clinical trials and show to outperform tumor assessment only, it may become a powerful decision-making parameter useful for dose escalation or combined modality trials.

Keywords: FMISO; tumor and lymph node hypoxia

Permalink: https://www.hzdr.de/publications/Publ-24110
Publ.-Id: 24110