Treatment outcomes for HPV-associated tonsillar cancer: a single-institution experience of 374 cases
Joongyo Lee,
Korea Republic of
PO-1109
Abstract
Treatment outcomes for HPV-associated tonsillar cancer: a single-institution experience of 374 cases
1Yonsei Cancer Center, Radiation Oncology, Seoul, Korea Republic of
Show Affiliations
Hide Affiliations
Purpose or Objective
Human
papillomavirus (HPV)-associated tonsillar cancer has a better prognosis than
HPV-negative cancer, so deintensification strategies to reduce or exclude
radiotherapy (RT) have been suggested through several studies. However, there
is no strong evidence or guidelines for deintensification of RT. In this study,
we investigated the treatment outcome in patients with HPV-associated tonsillar
cancer and provide deintensification strategy for RT.
Material and Methods
Retrospective
cohort study of patients with clinical stage T1-4N0-3 HPV-associated tonsillar
cancer treated between 2008 and 2020 with primary surgery or RT. Overall
survival (OS), progression-free survival (PFS) and cumulative incidence of locoregional
failure (LRF) between primary surgery and primary RT were analyzed, and propensity
score matching was performed to adjust for clinical factors. The following
subgroup analysis was conducted for patients who received primary surgery; The
difference in LRF according to adjuvant RT, prognosis differences according to
pathological response after neoadjuvant chemotherapy, and risk factors related
to contralateral regional failure in initial contralateral neck lymph node (LN)-negative
patients.
Results
Of
the total patients, 84 patients (22.5%) received primary surgery alone, 224
patients (59.9%) received primary surgery plus adjuvant RT, and 66 patients (17.6%)
received primary RT. After adjusting for clinical factors, there was no
statistical difference in OS, PFS, and LRF between the primary surgery group
and the primary RT group. In subgroup analysis, advanced pathologic N stage,
contralateral LN metastasis at diagnosis, abutting
or positive surgical resection margin, and no adjuvant RT were independent risk
factors of LRF in patients undergoing primary surgery. There was no
locoregional failure or death among 22 patients who had received neoadjuvant
chemotherapy before surgery and achieved pathological complete remission. Among
them, 15 patients (68.2%) did not receive adjuvant RT. Among 282 initial
contralateral neck LN-negative patients who underwent primary surgery, 8
patients had contralateral regional failure. None of these patients underwent
contralateral neck dissection, and all resection margins were less than 1 mm or
positive. In multivariate analysis, lymphovascular invasion and elective
contralateral neck irradiation not higher than 30.6 Gy were independent risk
factors of contralateral regional failure.
Conclusion
In
patients undergoing primary surgery, adjuvant RT can reduce LRF, and can be
further considered, especially for advanced N stage or abutting or positive
resection margin. Contralateral regional failure occurred in only 2.8% of initial
contralateral neck LN-negative patients who underwent primary surgery. However,
for patients who have adverse features such as lymphovascular invasion and
close resection margin and have not undergone contralateral neck dissection,
elective contralateral neck irradiation with a dose higher than 30.6 Gy can be
considered.