No consensus has yet been reached nor any robust evidence is available on the benefits of contralateral neck irradiation for patients with contralateral nodal-negative OSCC. Furthermore, RT-associated acute and long-term toxicities are significantly impacted by the treatment volume in patients with head and neck cancer, especially those receiving trimodality therapy (radical surgery plus concurrent chemoradiotherapy). With improved radiation technique, IMRT allows to increase the tailoring treatment volume to maximize the covering field while limiting the doses to normal tissues. As radiation oncologists aim to achieve a balance between the radiation toxicity and treatment outcome, the chance to omit the unnecessary treatment field without compromising the cure rate needs immediate investigation.
Recent retrospective data reported a low contralateral neck recurrence rate in patients with head and neck cancers. Table 3 lists a literature review of contralateral neck failure rate in primary head and neck cancers, particularly cancer at the oral cavity sites 11–15, 18,19. Unlike most of the historical analysis of mixed head and neck cancers and relatively early-stage OSCC, our study focused on a homogeneous patient series with well-lateralized and a more advanced-stage OSCC to avoid confounding interactions between different origins of head and neck cancers. According to the results, this study demonstrates a comparable low cRF to patients with high consistency advanced-stage OSCC.
In our cohort, 5-yr cRF was as low as 3.57%. No cRF was identified in patients who received neck irradiation, either UNI or BNI. Hence, omitting the contralateral N0 neck might be a reasonable approach in such patients. Vergeer et al. investigated 123 patients with oral cavity cancer (85%) and oropharyngeal cancer without contralateral neck irradiation and reported a cRF rate of 5.7% 18; however, their group included 7% of patients with close/cross midline disease, which could possibly increase the cRF. Wirtz et al. reported a cRF rate of 6.1%, mainly in the oropharynx (52.8%), with extended data for oral cavity (38%) and hypopharynx (10%) 12; however, 73.1% of their patients received contralateral neck dissection, which could have been an overtreatment. Another phase II study demonstrated a low cRF rate of 2.8% for resected head and neck cancers. Similarly, 71% of patients had a cross-midline disease and up to 92% of patients received bilateral neck dissection before adjuvant RT 15. Notably, our cohort, comprising 79.2% of patients with the T3/4 tumor, still revealed a reliably low contralateral neck recurrence rate, even though nearly 90% of them did not undergo contralateral neck dissection. Although contralateral neck dissection could be actively performed to detect occult contralateral nodal metastasis, there were certain short-term and long-term impacts on the quality of life of the patients 20.
Tumor-dependent factors that predicted contralateral neck recurrence are not well known and are still under investigation. The Sentinel European Node Trial included patients with lateralized, early T, and N0 tumors and demonstrated that the positive contralateral sentinel node was detected in only 1.9% of cases, but in up to 6% of cases of midline tumor 21. Al Mamgani et al. found midline involvement as the most significant correlation with cRF in a pooled analysis (12.12% with midline involved vs. 1.71% with free midline, P = .001) 22. It is generally regarded that there is a high frequency of lymphatic vessels crossing over the midline in certain tumor localizations (e.g., part of the oral cavity, tongue, and floor of the mouth) 23. ENE is also regarded as a predictor of cRF. Two large retrospective studies identified ENE as a strongly independent risk factor for 5-yr cRF [HR: 12.978, 95% CI: 1.328–126.829, P = .028] and for cRF in patients showing local recurrence (HR: 4.957, 95% CI: 1.763–13.934, P = .002) 8,9. As a result, one should be very cautious before deciding to omit contralateral neck irradiation in cases of midline crossing primary and ENE-positive status cancers. However, they were initially excluded from our study cohort because midline crossing primary and ENE-positive status cancers usually drive more extensive treatments (e.g., elective contralateral neck dissection and/or prophylactic contralateral neck field irradiation) by clinical physicians. A prompt discussion should be conducted at a multidisciplinary tumor board when treating such patients with N0 neck.
Though the univariate analysis conducted herein did not show any significant clinicopathological predictors for cRF, some cRF predictors have been previously reported (Table 3). Vergeer et al. showed that the number of nodes involved in the ipsilateral neck is a prognostic factor for cRF 5. Hence, a higher N stage might impact the cRF rate. Liu et al. reported that a tumor depth of invasion (DOI) of > 10 mm is a significant predictor in small (T1–2) lateralized OSCCs (HR: 6.7, 95% CI 1.4–32.3, P = .02) 11, which indicates that a higher T stage from patients diagnosed with early T1/2 in the AJCC 7th staging system can be revised to T3/4 in the DOI-incorporated edition of the AJCC 8th staging system. It might also impact the cRF rate. As our study mainly used the AJCC 6th and 7th staging systems, some T1-2N0 cases, which were earlier excluded from our analysis, might get upstaged if the AJCC 8th staging system is used. However, except for the influence of DOI, there were still no factors significantly associated with cRF in our study (which mainly included nearly 80% of T3/4 lesions), possibly owing to fewer events of cRF.
Notably, contralateral neck recurrence was detected as the first site of failure in five patients in our cohort, all of whom also simultaneously exhibited local recurrence. No isolated contralateral neck recurrence was observed. The failure pattern between the local failure and the cRF of other studies has also been reviewed in Table 3. Wirtz reported simultaneous local and contralateral regional failure in 42% of patients with cRF 12. Contreras reported that half of cRF occurred simultaneously with local failure and the other half occurred following the local failure 15. Though these studies did not conduct any statistical tests of the correlation between local failure and cRF, the results implied that contralateral neck should be more closely examined or followed up when local failure was found. It is generally considered that patients with a history of dissection or RT in the neck may have an aberrant lymphatic drainage caused by the disruption of lymphatic channels. This concept was verified and tested by SLNB in a new study, given the result of unexpected drainage pattern variability in 30% of patients with cT1-2N0 OSCC 24. Hence, SLNB has become an emerging technique that can benefit the staging of the contralateral negative neck and avoid the overtreatment of the contralateral neck in the future 25. Achieving a better local control might still be a priority in clinical situations for patients with a low risk of contralateral nodal recurrence.
The study has several limitations. First, the study could have a selection bias because of its retrospective nature; patients with more advanced stage or risks are likely to receive more intense treatments such as bilateral neck dissection or irradiation. Second, distance of the primary tumor to the midline was likely not measured with standardization. Third, the multivariable analysis should be considered exploratory because of the low number of cRF events, which limits its statistical power. Furthermore, for a group of highly consistent subsites of OSCC, our results are only applicable to well-lateralized buccal, cheek, gum, and retromolar OSCC population; therefore, caution should be exercised before extrapolating the results to ENE positive, tumor cross midline, or other head and neck cancers. Nonetheless, this study shows a reliable low cRF rate in well-lateralized OSCC with locally advanced stage and provides support for the possibility of omitting treatment to the contralateral neck in this group. At present, we are designing a more rigorous study to prospectively validate these results.