ESTRO 2023

Session Item

Head and neck

Session Code: 6005

Session Type: Poster (Digital)

Track: Clinical

Sentinel lymph nodes identification with SPECT/CT for HNSCC radiotherapy: a multicentric dummy run

Jean-Francois Daisne, Belgium

Presentation Number: PO-1209

Abstract

Abstract Title:

Sentinel lymph nodes identification with SPECT/CT for HNSCC radiotherapy: a multicentric dummy run

Authors:

Jean-Francois Daisne^1,2, Stephanie Deheneffe³, Tatiana Dragan⁴, Frederic Duprez⁵, Heleen Bollen¹, Sandra Nuyts^1,2, Michel Martens⁶, An Vancleef⁷, Annelies Maes⁸, Mieke Govers⁸, Dirk Van Gestel⁴, Chris Bruyninckx¹

¹University Hospitals Leuven, Radiation Oncology, Leuven, Belgium; ²Leuven Kanker Instituut, Oncology, Leuven, Belgium; ³CHU-UCL-Namur (site Ste-Elisabeth), Radiation Oncology, Namur, Belgium; ⁴Institut Jules Bordet, Université Libre de Bruxelles, Radiation Oncology (Head and Neck Unit), Brussels, Belgium; ⁵University Hospital Gent, Radiation Oncology, Gent, Belgium; ⁶AZ Turnhout, Radiation Oncology, Turnhout, Belgium; ⁷Onze Lieve Vrouw Hospital, Radiation Oncology, Aalst, Belgium; ⁸Limburgs Oncologisch Centrum, Jessa Hospital and Ziekenhuis Oost-Limburg, Radiation Oncology, Hasselt and Genk, Belgium

Show Affiliations

Purpose or Objective

In the multicentric, prospective phase 2 randomized SEMIRAHN study (NCT04688528), sentinel lymph nodes (SLNs) are identified on 3-dimensional single photon emission computed tomography (SPECT)/computed tomography (CT) images to individually tailor the elective nodal irradiation (ENI) volume of head and neck squamous cell carcinomas (HNSCC). As a part of the mandatory patient-based radiotherapy (RT) dummy run, each center must handle SPECT/CT images. The aim is to test the ability to read, interpret and transfer the SLN drainage information to the selection and delineation of the ENI clinical target volume (CTV).

Material and Methods

The anonymized data of a patient with a cT1N0M0 left hypopharyngeal SCC, treated in a previous RT study with SLN identification for ENI, were securely sent to the 7 participating centers. It encompassed the description of the clinical case, the description of the endoscopic examination, the simulation CT slices after iodine contrast injection and the hybrid SPECT/CT images after 99mTc-labelled nanocolloid endoscopic injection around the tumor and migration into the SLNs. Each local radiation oncologist (RO) and nuclear medicine specialist were asked to process the SPECT/CT images on the local infrastructure and to report all identified SLNs by decreasing order of activity and localization in the neck (level and laterality). Each RO was asked to delineate a mandatory list of organs at risk, the primary CTV and the ENI CTV (named “CTVn-LS”) encompassing the nodal levels containing up to the 4 hottest SLNs on the slices of the simulation CT. We report here the results of the SPECT/CT reading and its impact on the selection of the CTVn-LS levels.

Results

No center reported problems to read the SPECT/CT images. Four SLNs were identified by 5 centers (in levels 2 right (R), 2 left (L), 3L and 2L; by decreasing order of activity). Two centers did not identify the SLN in level 2L showing a low activity (about 10% of the SLN in level 2R with maximal activity, and about twice the background activity). All centers selected levels 2L, 3L and 2R for the CTVn-LS, and 1 center added level 5L (because of 3rd SLN in level 3L at its very posterior edge).

Conclusion

All centers could Identify and locate the 3 most active SLNs on the SPECT/CT images, and transfer the information to delineate the CTVn-LS on the slices of the simulation CT. One SLN with a very low activity was missed by 2 centers, showing that images saturation display and/or subjective interpretation of the images play a role.