The PP NWZ Alkmaar, has a weekly clinic and MDT for HNC patients. The new patients are examined by a head-neck surgeon (otolaryngologist and/or maxillofacial surgeon) and a radiation oncologist. A dermatologist is consulted when necessary. The diagnostic workup was usually performed within two weeks and the patients returned to the clinic for results, after discussion with their MDT. The PP has the same equipment as the HNOC and used the HNOC imaging protocols throughout the study. The patients were referred to the HNOC, during the following week.

The Head and Neck Oncological Center, Amsterdam UMC, location VUMC, has a weekly clinic and MDT for HNC patients. All available information was reviewed, and additional information was gained. All radiologic studies are send to the HNOC, for formal re-evaluation by the head and neck radiologists of the HNOC. The patients are seen in the outpatient clinic, where a complete history is done. Standard head and neck examination includes: inspection of the oral cavity and pharynx bimanual palpation of the mouth, palpation of the neck, (in)direct laryngoscopy and stroboscopy (in patients with a small laryngeal carcinoma).

A management proposal was formulated prior to this MDT by the PP MDT. During the MDT, head and neck surgeons, radiation and medical oncologists were present, besides the radiologists and pathologists, other physicians on indication and supporting staff. The PPs head and neck surgeons and radiation oncologist are attending the HNOC MDT via teleconference. During the MDT discussions MR were based on the current NCCN and national guidelines. Management recommendations are approved or changed and reported back to the PPs physicians. The final MR of the HNOC MDT was usually carried out in the PP, Patients with a nasopharyngeal carcinoma or a skull base malignancy were treated in the HNOC, due the relative rarity in our population. The PP has the possibility to perform many head and neck surgical procedures including microlaryngeal laser surgery, microvascular reconstructions, and to treat patients with (chemo)radiotherapy. In the time frame of the present study the surgeons at the PP, where trained to perform sentinel node procedures. Therefore in the first 6 years of the study patients were referred to the HNOC for treatment, and in the latter 3 years patients were treated in the PP. Total laryngectomies are only performed at the HNOC.

After gaining approval of the institutional review boards in the HNOC, Amsterdam UMC, location VUMC, data was collected and scored using Microsoft Excel (Microsoft, Redmond, Washington). Scoring was done by three independent individuals (JJHT, TM and JMU), of which two are senior attendees of the HNOC MDT (JJH and JMU) to reduce bias. Statistical analysis was performed using SPSS version 22.0 (IBM Corporation, Armonk, New York). Descriptive statistics were used for all data. To compare categorical data, the Chi-square or Fischer’s exact test were used, where a value of p < 0.05 was considered statistically significant.

Between January 2011 and October 2018, 548 patients were referred by the PP to the HNOC. In the weekly HNOC-MDT, 10-15 new patients were discussed each week; therefore, during this period, the HNOC-MDT discussed approximately 4500 patients. Thirty-three patients were excluded due to the following reasons: only discussion in the PP-MDT (n = 17), unspecific MR by the PP (n = 10), patients referred with benign head and neck tumors (n = 6) (see Figure 1 ). Twenty-seven patients were referred more than once; 25 patients among them were referred twice, and one patient was referred three times. Resulting in 515 evaluable MDT recommendations.

The clinical data is summarized in Table 3 . The mean age of the patients at the time of the MDT was 65 years, and 65% of the patients were male. The most common tumor sites were the oral cavity (n = 185), larynx (n = 129), oropharynx (n = 89), hypopharynx (n = 29), nasal cavity (n = 27), salivary glands (n = 23), unknown primary tumor (n = 19), nasopharynx (n = 3), and others (n = 11). The tumor sites defined as ‘other’, included five cases of cutaneous squamous cell carcinomas and a cutaneous melanoma, squamous cell carcinoma of the external auditory canal, cutaneous sarcoma, Merkel cell carcinoma, recurrent papillary thyroid cancer with tracheal extension and a neck metastasis of a cutaneous squamous cell carcinoma. Over 85% of the cases were primary tumors. Recurrent or residual tumors were seen in 7% of cases and second primary tumors were discussed in 5.6% of the cases. There were no cases with a distant metastatic tumor.

TNM-classification was missing in 46 cases (14 primary, 3 second primary, 25 recurrent and 4 residual tumors).

The most commonly recommended treatment by the PP, was surgery with or without adjuvant treatment based on subsequent histopathological findings (n = 172), followed by primary radiotherapy (n = 114), surgery with planned adjuvant radiotherapy (n = 83), chemoradiotherapy (n = 95) and chemotherapy (n = 1). Some patients of the PP were referred with two or more MRs, eg. Primary chemo-radiotherapy or upfront neck-dissection with adjuvant (chemo-)radiotherapy (n = 50). During the study population several studies were conducted in both the PP and in the HNOC. Inclusion in studies, only ongoing in the HNOC, were not scored as a MR-change.

Change in MR occurred in 113 cases (22%) and major changes occurred in 86 cases (16.7%). In 67 (59%) of these cases, more intense treatment was recommended (e.g. performing a bilateral neck dissection, where a unilateral neck dissection was proposed). In 43 cases (38%), de-intensification (e.g. only radiotherapy, where concomitant chemotherapy was proposed and in three cases (2%) a different recommendation with the same intensity was proposed. In 63 (12.2%) of the cases the HNOC-MDT advised to perform additional diagnostics e.g. a second ultrasound-guided fine needle aspiration to assess possible regional metastasis. However these recommendations did not result in a change in MR.

Likewise in 63 cases (12.2%), there were minor recommendations that caused changes in tumor re-classification, without an change in MR. E.g. the change from a cT1 oral cavity to an cT2 oral cavity, for which still a transoral excision and sentinel node procedure was recommended.

In 49 cases, imaging review caused minor changes in the recommendation and in the case of 19 patients, this was due to histopathology review. See Table 4 for details.

Physical reexamination changed MR in 21 cases (4%) (see Table 5 ). Major changes occurred in 18 cases (3.5%) Adjusted classification following physical examination occurred in 22 cases (4%), and histopathology samples were reviewed in 231 cases (45%) by dedicated head and neck pathologists. In 21 cases, the histopathology diagnosis changed after revision (9.1%). This change of diagnosis in 11 cases (52%) had a direct impact on the MR. Additional histopathology was requested in 38 cases by the HNOC. In the HNOC, a dedicated head and neck radiologist revised the diagnostic imaging. Changes in imaging reports were found in 129 cases (25%). These changes in 37 cases (28%) had a direct impact on the recommended management. The TNM-classification was adjusted in 95 cases (21%) after the HNOC MDT, and this was most frequently done because of changes in diagnostic imaging reports (71%) and diagnostic physical evaluations (23%) (see Table 6 ), 63% were down-classified. In 32 cases (15% of total number of cases discussed) both TNM-classification and MR changed. Additional imaging was requested in 93 cases (18%). In 45 cases (8.7%), an additional endoscopy under anesthesia was performed in the HNOC.

Almost half (46%) of the cases referred to the HNOC had a change in either diagnostics and/or MR. Changes in MR after the HNOC-MDT were found in 22% of the cases, with major changes in 16% of the cases, such as bilateral instead of one-sided neck dissections or the addition of a treatment modality such as (chemo)radiotherapy. The rate of changes in MR during MDT’s in other oncologic fields show a wide range in the literature, ranging from 2% up to 27% (19, 21–23). These differences could be attributed to differences in the validity of pre-MDT recommendation as well as the differences in the details of the recommendation studied (e.g. modality only, drug type, dosage, etc).

A similar collaboration between the University Medical Center Groningen (HNOC) and the Medial Center Leeuwarden (PP) was investigated (24). In this prospective study, eight cases out of the 336 cases (2%) involved a change of recommendation. It should, however, be noted that the patients were presented via video-conferencing. Whereas patients were physically examined in the HNOC in the present study. Furthermore, the radiologic investigations were presented via video-conferencing, whereas the radiologic investigations were completely transferred and formally re-evaluated in our collaboration. Additionally, the pathology slides were revised on indication only. In our series there was no selection bias since all patients were re-evaluated.

The investigated outcomes were changes in MR. The present study does not allow for oncological outcome analysis since all patients were treated according to the recommendations of the HNOC-MDT. Making comparison of patient with and without changes in recommendation difficult to interpret. Comparing oncological results of the current population with those from the literature or national databases was thought to be problematic because of different patient populations and given the fact that the literature usually reports a variability and range in survival.

Treatment intensification is usually instituted with the aim to improve survival, e.g. concomitant chemoradiotherapy vs. Radiotherapy improves survival by 6.5% (25). Moreover treatment deintensification, when appropriate, reduces complications and side-effects, without jeopardizing survival (26). There is considerable evidence that patients treated in centers with a high volume have a better survival (27). These aspects combined suggest that the oncological outcome likely will have been improved by the recommended changes.

There was no association between changes in recommendations and certain subgroups. The implication of this is that all patients may benefit from referral to a high volume MDT.

Several studies have found that the revising of histopathological slides affects MR. Kronz et al. (28) analysed the literature and found that management changed in 5-7% of the cases because of a second opinion and evaluation of pathology. Westra et al. (29) found that in 7% of the patients, major discrepancies were identified after the second opinion pathology evaluation for HNC. After examining multiple organ systems, a prospective review found that 1.4% of the 6000 cases witnessed a change of diagnosis due to second opinion pathologies (30). Tung et al. (31) found that in 6% of the 715 cases a major change in pathology occurred and in 2% of the total cases, the management was changed after the second opinion pathology. Garcia et al. (32) found that in 20% of cases the revised pathology led to a change in pathology diagnosis. Comparing the aforementioned literature to our study, we found management changes on the basis of revision histopathology in 5% of cases. Since difficult histopathological cases of the PP are sent to the HNOC pathology department for revision independent of the reevaluation in the MDT, this practice likely lead to an artificially low figure.

For diagnostic imaging in HNC, Lysack et al. (33) investigated the impact of second opinion diagnostic imaging review by a dedicated head and neck radiologist. They found an adjusted TNM-classification in 56% of cases, and MR changed in 38% of these patients. Hatzogou et al. (34) found that second-opinions in neuroimaging, resulted in disagreement in 55 (19%)patients, and the management or disease classification was changed in 39 (15%) patients out of 55. Recently the importance of the presence of an head and neck radiologist was investigated by Alterio et al. (35), showing that in 52% of the cases there was a change to the initial radiologic report. In 27% the imaging was considered in adequate. And in 25% there was a modification in tumor staging or in MR. When comparing this with our population, adjustments in classification could be attributed to radiology solely in 25% of the cases. However, this resulted in a change of MR in only 7% of the cases. Therefore, the absolute percentage of radiology adjustments aligned with the literature.

While examining the effect of the second evaluations on classification, we found changes in 21% of the cases. The study of Wheless et al. (12) found in changes in classification in 12% of cases. Bergamini et al. (18) found reclassification necessary in 49% of the cases, without stating further details. In the second study, requests for better classification were registered instead of actual changes. Explanations for the change in classification could be the time interval between examinations, combined with the rate of progression (36). Although the time between the two MDT recommendations was not measured in the present study, the interval was likely to a maximum of 3 weeks since national guidelines were strictly adhered to.

The current study was retrospective in by nature, which was both its strength as well as its limitation. As such there was no control group without the HNOC-MDT. Therefore a direct comparison fo the HNOC-MDT influence on outcome was not possible.

Furthermore the retrospective nature, caused difficulties to determine the relative effect of physical examination/pathology/radiology on the changes in MR.

In this study, the HNOC-MDT is considered the gold standard. This is based on the higher volume of patients and superior experience of the team members.