2018-19 Section 7-Neoplastic and Inflammatory Diseases of the Head and Neck eBook

Oral Oncology 73 (2017) 152–159

K.Y. Zhan et al.

Fig. 3. Kaplan Meier survival plot of AJCC 8th edition nodal staging.

Number at Risk 12 months

24 months

36 months

48 months

pN0 pN1 pN2

272

197

100 811 134

43

1848

1404

365

310

218

59

recently reported the signi fi cant negative survival e ff ect of ENE in HPV + OPSCC using the NCDB, an e ff ect that persisted when restricting their analysis to patients with one positive lymph node [19] . Moreover, they did not fi nd a signi fi cant survival di ff erence in ENE+ patients with adjuvant chemoradiation vs. adjuvant radiotherapy alone. Similarly, Kumar et al. demonstrated that adding post-operative chemotherapy to radiotherapy in surgically-managed HPV+ OPSCC with ENE did not confer a survival advantage [20] . The above fi ndings contrast smaller, single institution studies that have not shown prognostic e ff ect of ENE in HPV+ OPSCC [14 – 16,21,22] . Haughey et al. in their multi-institu- tional trial (n = 704) found ENE to have a hazard ratio of 1.61 (95% CI 0.98 – 2.63, p = 0.06). Given the moderate e ff ect size on survival with ENE (5 – 11% on 4-year OS in our series), it is not surprising to see this signi fi cant fi nding only in larger, registry-based studies. As our knowledge of tumor biology and molecular genetics improves, ENE could be incorporated into staging for further hazard discrimination and numerical balancing. Nevertheless, AJCC 8th edition provided signi fi cant staging improvements moving forward. Limitations to this study include its retrospective nature and het- erogenous cohort of HPV+ OPSCC patients, which includes unknown primaries and patients treated with both adjuvant chemotherapy and/ or radiotherapy. The NCDB records only timing of primary site surgery with other treatment modalities, obfuscating the timing of primary treatment modality with cervical lymphadenectomy, and whether sur- gery was for primary treatment vs. persistent disease. Our selected group of surgically-treated patients confers an unknown degree of treatment selection bias for surgically-favorable patients and disease. ENE is not systematically de fi ned in the coding manual as this variable is simply abstracted from pathology reports. Finally, we cannot com- ment on the degree of environmental exposures such as smoking or

alcohol use, as these data are not recorded in the NCDB. Nonetheless, our updated staging cohort closely approximates that of contributory studies to the AJCC 8th edition [3,4,8] . More importantly, our national sample consists of patients treated at both academic and non-academic settings – a more representative cohort with more heterogenous treat- ment approaches. Including unknown primaries is also bene fi cial given that this has been incorporated into the overall staging of these tumors as pT0. This study has demonstrated the dramatic shift in staging from AJCC 7th to 8th edition for HPV+ OPSCC using a large national sample, the vast majority patients now reclassi fi ed as pStage I. Our data shows far improved hazard discrimination and outcome prediction in the AJCC 8th edition ’ s overall pathologic staging compared to the 7th. Nonetheless, pathologic nodal staging alone does not demonstrate se- parate survival curves with the exception of pN2. A potential area of further strati fi cation is pathologic ENE, whose presence confers a modest but statistically signi fi cant decrease in overall survival. However, further study of ENE and strati fi cation are needed.

Author ’ s note

The NCDB is a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The data used in the study are derived from a de-identi fi ed NCDB fi le. The American College of Surgeons and the Commission on Cancer have not veri fi ed and are not responsible for the analytic or statistical metho- dology employed, or the conclusions drawn from these data by the investigator.

181