To investigate the feasibility and accuracy of an automated method to validate gross tumor volume (GTV) delineations with pathology, in laryngeal and hypopharyngeal cancer. High resolution CT (CTHR), MRI, and positron emission tomography (PET) scans were obtained from 16 patients before total laryngectomy. The GTV was delineated in each imaging modality, separately. The laryngectomy specimen was fixed with formaldehyde, embedded in a block of agarose, sliced transversely in 3-mm thick slices, photographed, and whole-mount hematoxylin-eosin stained sections were obtained. Tumor tissue was delineated in the hematoxylin-eosin sections by a pathologist. A landmark based registration was performed to register the hematoxylin-eosin sections to the thick-slice photos. Using the thick-slice photos, an automatic 3D reconstruction of the specimen was performed. The CTHR was automatically and rigidly registered to the 3D-specimen. The MRI and the PET were rigidly registered to the 3D-specimen using the CTHR as intermediate. The accuracy of the pathology-imaging registration and the specimen deformation and shrinkage were assessed. The tumor delineation inaccuracies were assessed as the distance between the surfaces of the imaging based GTVs and the gold standard delineation (based on the hematoxylin-eosin sections). The tumor delineation inaccuracies were finally compared with the registration errors, to estimate the limits of the validation of the GTV delineations with our registration method. Good agreement was observed between anatomical landmarks in the 3D-specimen and in the in vivo images. Limited deformations and shrinkage (3 ± 1%) were found inside the cartilage skeleton. However, deformations were observed outside the skeleton, particularly in the epiglottis. The root mean squared error of the registration between the 3D-specimen and the CT, MRI and PET was on average 1.5, 3.0, and 3.3 mm, respectively, in the cartilage skeleton. The GTVs delineated based on the hematoxylin-eosin sections, CT, MRI, and PET generated a mean volume of 7.2, 14.9, 18.3, and 9.8 mL, respectively. The tumor coverage by the CT, MRI, and PET delineation was on average 85%, 88%, and 77%, respectively. The tumor delineation inaccuracies exceeded the registration error for 62%, 38%, and 37% of the GTV surface points, for CT, MRI, and PET, respectively. Validation of GTV delineation with pathology is feasible with an average overall accuracy below 3.5 mm inside the laryngeal skeleton. The tumor delineation inaccuracies were larger than the registration error; therefore, an accurate histological validation of anatomical and functional imaging techniques for GTV delineation is possible, in laryngeal and hypopharyngeal cancer patients.