- Email: [email protected]
Patterns of failure in high-metastatic node number human papillomavirus-positive oropharyngeal carcinoma
Oral Oncology 85 (2018) 35–39
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
Patterns of failure in high-metastatic node number human papillomaviruspositive oropharyngeal carcinoma
T
Nicholas C.J. Leea, Jacqueline R. Kellya, Henry S. Parka, Yi Ana, Benjamin L. Judsonb, ⁎ Barbara A. Burtnessc, Zain A. Husaina, a
Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, USA Section of Otolaryngology, Department of Surgery, Yale School of Medicine, New Haven, CT, USA c Section of Medical Oncology, Department of Internal Medicine, Yale School of Medicine, New Haven, CT, USA b
A R T I C LE I N FO
A B S T R A C T
Keywords: Head and neck cancer Oropharyngeal carcinoma Human papillomavirus Distant metastasis Radiotherapy Surgery Neoplasm Oropharynx Squamous cell cancer
Background: The 8th edition American Joint Committee on Cancer staging system for resected HPV-positive oropharynx carcinoma (HPV+ OPC) highlights high node number as a critical determinant of survival. We sought to characterize outcomes and patterns of failure in patients with high pathologically involved node number oropharynx cancer. Methods: We retrospectively identified 116 HPV+ OPC patients sequentially treated with neck dissection and either resection or intraoperative brachytherapy of the primary tumor between 2010 and 2016. External beam radiation was given based on the pathologic findings. Cox proportional hazards regression was used for multivariate analysis. Results: With a median follow-up of 27 months, the 3-year overall survival and progression free survival (PFS) were 89% and 81%, respectively. On multivariate analysis, ≥5 involved lymph nodes was significantly associated with worse PFS (hazard ratio 4.3, 95% confidence interval (CI) 1.5–12.0, P = 0.001). Rates of 3-year locoregional recurrence (LRR) in patients with ≤4 vs ≥5 were 6% and 22% (log-rank P = 0.12). Rates of 3-year distant metastases (DM) were 12% and 53% between ≤4 and ≥5 (log-rank P < 0.001). Conclusion: Our findings confirm that patients with 5 or more involved lymph nodes appear to have substantially worsened rates of disease recurrence. While these patients appear to be at high risk of both LRR and DM, the predominant mechanism of failure is distant, and the rate of DM in this group was over 50%. Dedicated clinical trials in this patient population are warranted with a focus on mitigating the high DM rate.
Introduction Human papillomavirus-positive (HPV+) oropharyngeal carcinoma (OPC) compromises over 70% of all newly diagnosed OPC cases and exhibits more favorable survival outcomes compared to HPV-negative disease. Survival rates with HPV+ disease are approximately 30% higher at 5 years than with HPV-negative disease [1], which has led to the design of different staging systems for both HPV+ and HPV-negative disease. Additionally, amongst HPV+ patients, a separate staging system has been adopted for clinically staged versus pathologically staged patients [2]. In the new pathologic staging system, the number of involved nodes, as opposed to laterality or size, is a key determinant of stage in resected cancers. In the study that formed the basis for the
American Joint Committee on Cancer (AJCC) 8th edition staging system, the presence of ≥5 pathologically positive nodes was associated with worsened 5-year overall survival (OS) 71% vs 84% [3]. Yet, the patterns of failure within this high-risk group were not specifically elucidated—an important aspect for developing therapeutic strategies to counteract the worsened survival rates. We sought to further examine outcomes within a cohort of patients treated surgically with neck dissections and either resection or intraoperative brachytherapy of the primary site to examine risk factors for recurrence, validate the ≥5 node cutoff, characterize the role of salvage therapy in patients with recurrence, and, most importantly, examine patterns of failure in high (≥5) node number versus low (≤4) node number patients.
Presented in part at the 2017 American Society of Therapeutic Radiation Oncology Conference (September 26, 2017; San Diego, CA) and 2018 American Radium Society Conference (May 8, 2018; Orlando, FL). ⁎ Corresponding author at: Department of Therapeutic Radiology, Yale School of Medicine, 35 Park Street, New Haven, CT 06511, USA. E-mail address: [email protected] (Z.A. Husain). https://doi.org/10.1016/j.oraloncology.2018.08.001 Received 12 April 2018; Received in revised form 18 July 2018; Accepted 1 August 2018 1368-8375/ © 2018 Elsevier Ltd. All rights reserved.
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
Methods
Table 1 Patient demographics.
Patient population HPV-associated OPC patients from 2010 to 2016 at Yale New Haven Hospital (YNHH) who underwent neck dissection were included in this retrospective study. Those with previous head and neck cancer or cancer treatment to the head and neck region were excluded. Yale University’s IRB approved the protocol and written consent was waived. Pathologic data, HPV status, and clinicodemographic features were extracted from patient records. HPV status was determined using p16 immunohistochemistry and/or HPV in situ hybridization (ISH). Study endpoints Progression free survival (PFS) was defined as time from primary surgery to progression or death from any cause. OS was defined as time from primary surgery to death of any cause. Distant metastases (DM) were defined as disease outside of the head and neck region and was confirmed by imaging and/or biopsy-proven disease consistent with the primary when possible. Locoregional recurrence (LRR) was defined as recurrence within the head and neck. Statistical analysis Statistical analysis was performed using STATA/IC software package (version 14.2; StataCorp, College Station, TX). Multivariable analysis (MVA) was conducted using backwards-exclusion with the Cox proportional hazards model. The inclusion cut-off was P < 0.20. Proportionality was confirmed using time-dependent covariates. Survival analysis was conducted using the Kaplan-Meier method and log-rank test. Results Patient and tumor characteristics One hundred and sixteen patients were included with a median follow-up of 27 months (range: 1 to 63 months). Patient and tumor characteristics are described in Table 1. The median patient age was 60, and 101 (87%) patients were male. 64 (55%) of patients had a tonsil primary while 52 (45%) had a base of tongue primary. Ninety-eight (85%) were cT1-2, and 83 (72%) were 7th Ed. AJCC pN1-2b. 11.2% had greater than 4 pathologically involved lymph nodes (range 0 to 14). The median number of lymph nodes dissected was 34 (range: 5–97). Fiftynine patients (51%) had pENE. Forty-one patients (35%) received intraoperative brachytherapy (IOBT) to the primary site without resection while 75 patients (65%) received primary tumor resection. Of the resected patients, 35 underwent transoral robotic surgery (TORS). There was a strong trend for patients undergoing primary brachytherapy to having more advanced T stages than patients undergoing primary surgery (Supp. Table A). Ten patients (24%) receiving IOBT were cT3-4 while eight patients (11%) receiving primary tumor resection were cT34 (p = 0.051). Patients undergoing IOBT were treated with palladium103 with a median dose of 25.74 Ci (range: 17.75–44.4 Ci). Of those receiving resection, 11 had positive margins (14%). Overall, 64 patients (55%) received post-operative chemoradiation (CRT), 38 (33%) received adjuvant RT alone, and 13 (11%) received no post-operative therapy. Within those receiving no post-operative therapy, eight were Stage I-II and pN0 on neck dissection. Of those receiving chemotherapy, 48 (74%) received cisplatin, six (9%) received carboplatin, four (6%) received a combination, and eight (12%) were not definitively specified due to treatment at an outside facility. Adjuvant RT was delivered with IMRT at a median dose of 58 Gy (range: 50–70 Gy) in 30 fractions (median; range: 20–38). Amongst the 63 patients with radiation fields that were available for review, 59 patients received radiation to the
Characteristic
Total (%)
≤4 pN+ (%)
≥5 pN+ (%)
P
Total Age: median [range], y
116 60 [36–87]
103 (88.8) 59 [36–87]
13 (11.2) 62 [48–81]
0.14
Sex Male Female
101 (87.1) 15 (12.9)
88 (85.4) 15 (14.6)
13 (100) 0 (0)
Disease Site Tonsil and Soft Palate Base of Tongue
64 (55.2) 52 (44.8)
58 (56.3) 45 (43.7)
6 (46.2) 7 (53.8)
Smoking Never smoker Former smoker Current smoker
45 (38.8) 40 (34.5) 31 (18.7)
39 (37.9) 37 (35.9) 27 (26.2)
6 (46.2) 3 (23.1) 4 (30.8)
Smoking dose, pysa ≤10, Light > 10, Heavy
62 (54.9) 51 (45.1)
53 (51.5) 47 (45.6)
9 (69.2) 4 (30.8)
cTb T1 T2 T3 T4
40 (34.5) 58 (50.0) 12 (10.3) 6 (5.2)
37 (35.9) 50 (48.5) 10 (9.7) 6 (5.8)
3 8 2 0
(23.1) (61.5) (15.4) (0)
pN (AJCC 7th)c N0 N1 N2a N2b N2c N3
16 (13.8) 10 (8.6) 24 (20.7) 49 (42.2) 12 (10.3) 5 (4.3)
16 (15.5) 10 (9.7) 24 (23.3) 42 (40.8) 6 (5.8) 5 (4.9)
0 0 0 7 6 0
(0) (0) (0) (53.8) (46.2) (0)
Primary Treatment Brachytherapy Resection
41 (35.3) 75 (64.7)
33 (32) 70 (68)
8 (61.5) 5 (38.5)
pENE Negative Positive
59 (50.9) 57 (49.1)
49 (47.6) 54 (52.4)
10 (76.9) 3 (23.1)
Surgical margins Negative Positive
62 (86.1) 11 (13.9)
57 (83.8) 11 (16.2)
5 (1 0 0) 0 (0)
Adjuvant Therapyd None Chemotherapy RT CRT
13 (11.2) 1 (0.9) 38 (32.8) 64 (55.2)
52 (50.5) 1 (1.0) 38 (36.9) 12 (11.7)
12 (92.3) 0 (0) 0 (0) 1 (7.7)
0.21
0.49
0.71
0.27
0.68
< 0.001
0.06
0.08
> 0.99
> 0.99
y = year. pys = pack-years. cT = clinical T stage. pN = pathologic N stage. AJCC = American Joint Committee on Cancer. pN+ = number of pathologically positive lymph nodes. a Incomplete smoking data. b Comparison of cT1-2 v cT3-4. c Comparison of pN0-2a v pN2b-3. d Comparison of any adjuvant therapy to none. P values determined by Fisher’s exact test, chi-square, or student t-test where appropriate.
tumor bed and bilateral neck, while 4 patients received radiation to the tumor bed and ipsilateral neck only. Survival outcomes For all patients, the 3-year PFS and OS were 81% and 89% respectively. Of the 103 patients with ≤4 nodes, the 3-year PFS was 86%, while for the 13 patients with ≥5 involved nodes the 3-year PFS was 47% (Fig. 1; log-rank P = 0.002). However, 3-year OS was not significantly different in patients with ≤4 nodes at 90% and ≥5 nodes at 81% (log-rank P = 0.26). We found no significant difference in 3-year OS between patients with pENE (91%) and those without pENE (86%) (log-rank p = 0.30) or 3-year PFS (pENE+ 83% and 77%, log-rank P = 0.89). Among patients with resection of the primary tumor, there was no difference in 3-year 36
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
Fig. 1. Kaplan-Meier of Progression Free Survival by metastatic node number.
Notably, patients with ≥5 positive lymph nodes had a significantly higher rate of 3-year DM than patients with ≤4 nodes (53% versus 12%, p < 0.001).
OS between patients with positive versus negative margins. Patients had similar PFS, OS, and locoregional control (LRC) with IOBT or surgical management of the primary tumor (3-year PFS 74% v 80% logrank p = 0.22; 3-year OS 87% v 91%, log-rank p = 0.64; 3-year LRC 89% v 94%, log-rank p = 0.33, respectively). Additionally, there was no difference in 3-year OS based on 7th Ed. AJCC nodal staging (pN0-2a 89% v pN2b-3 88% log-rank p = 0.75). On univariate analysis of predictors of recurrence, two variables met criteria for inclusion in the PFS MVA: number of nodes dissected (continuous, P = 0.005) and the number of positive nodes (≤4 vs. ≥5, P = 0.001). After taking into account relevant clinicodemographic variables in our MVA, both were significant. The number of nodes dissected had a HR 1.04 and 95% CI 1.001–1.05. The lymph node threshold of ≥5 positive nodes remained highly significant (P = 0.005, HR 4.3, 95% CI 1.6–12.0) as demonstrated in Table 2.
Salvage therapy Five patients underwent salvage therapy with curative intent (median follow-up: 47 months; range: 23–69 months). Of these, two were patients with LRR and three were patients who recurred with oligometastatic disease that were treated with metastectomy and/or stereotactic body radiotherapy. Of the five patients, one patient had no evidence of disease (NED) at last follow up. This patient was originally a Stage IVB cT2N3M0 never-smoker who underwent upfront IOBT and bilateral neck dissection followed by external beam radiotherapy with cisplatin. Approximately one year and four months after finishing CRT, he was found to have an 8 mm left lower lobe (LLL) nodule on surveillance imaging. He underwent a salvage LLL metastectomy with pathology confirming two foci of p16-positive, HPV ISH-positive squamous cell carcinoma (8 and 7 mm). He is currently four years out from completion of treatment and remains NED.
Patterns of recurrence In total, seventeen patients (15%) developed disease recurrence (Table 3). Recurrences were detected by CT in six cases (35%), PET in ten cases (59%), and chest x-ray in one case confirmed by CT (6%). Biopsies were obtained in fifteen cases (88%), and all were consistent with the primary tumor. Seven patients developed locoregional recurrence: one was isolated, five were synchronous with DM, and one had LRR followed by DM. Of these seven LRR, five were ipsilateral and two were contralateral (both out of field). Of the ipsilateral LRR, three were in field, one out of field, and one did not receive RT. In addition, of the two patients who recurred in the neck, one was contralateral (in level III) and the other patient recurred at the same level of the initial disease presentation. Patients with ≥5 positive lymph nodes had a higher rate of 3-year LRR compared to ≤4 nodes, although this did not reach criteria for significance (22% versus 6%, respectively, log-rank P = 0.12). Sixteen patients developed DM: 10 were isolated, 5 were synchronous with LRR, and one occurred after an initial LRR. Of these, two patients developed disseminated metastases involving organs including the brain and testicles. The median time to recurrence was 1 year (range: 113 days–5 years). Three of our 16 patients (19%) who developed distant metastasis did so after at least two years of surveillance.
Discussion The 8th edition AJCC system for pathologically staged HPV+ OPC has made pathologic node number a key determinant of stage. This is based on work demonstrating that the AJCC 7th edition pN staging did not accurately stratify patients by oncologic outcome by survival or hazard analysis, findings that were also shown in our dataset [3,4]. In the multi-institutional study that became the basis for the 8th edition staging system, ≤4 LNs and ≥5 LNs were associated with 5-year OS values of 89% and 71%, respectively [3,5]. Here, the 3-year OS values were 90% and 81% for ≤4 LNs and ≥5 LNs, respectively, while the 3 year PFS values were 86% and 47% for ≤4 LNs and ≥5 LNs, respectively. The significant difference in PFS but not OS for our cohort is likely due to shorter follow-up compared to Haughey et al, which when combined with the pattern of late recurrences, likely did not allow for recurrences to convert into deaths [3]. A notable difference between our series and the Sinha et al. series is that while their experience also showed a four-fold increase DM (16% v 37
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
which helps inform therapeutic strategies for this high risk group of patients. Our work demonstrates that patients with 5 or more involved nodes are at risk for both LRR and DM, with values of 22% and 53% respectively. The most common site of failure for these patients is overwhelmingly distant. This finding is critical as it impacts the aim of clinical trials in this patient population. Currently, in ECOG 3311 (NCT01898494), all patients with 5 or more involved nodes receive concurrent cisplatin based XRT, with the idea being that these patients are at risk for LRR and adding radiosensitizing chemotherapy may be of benefit. This work suggests that in addition to a radiosensitizing agent for locoregional control, a focus on therapies that may affect the distant metastatic rate may be indicated. For example, concurrent docetaxel may be one such strategy. In RTOG 0234 (NCT00084318), a randomized phase II study, for high risk postoperative patients with HNSCC, compared cisplatin vs docetaxel with 60 Gy radiation [6]. This showed improved PFS in the docetaxel arm, largely due to a decrease in distant metastatic disease (2-year DM rate of 25% in cisplatin vs 13% in docetaxel). This strategy is being further investigated in the RTOG 1216 (NCT01810913) study, although it is aimed at a different population, namely high-risk (positive margin or pENE+), specifically excluding HPV+ oropharyngeal cancer patients. Alternatively, given the recent data suggesting a benefit to adjuvant immunotherapy in suppressing distant metastases in NSCLC after definitive local therapy, the ≥5 node group may represent a key demographic for adjuvant immunotherapy trials in HPV+ head and neck cancer [7]. Our data helps confirm other important findings in HPV+ OPC. Firstly, we demonstrate no significant difference in OS between patients treated with IOBT versus resection in HPV+ OPC, which is consistent with previous work in head and neck cancers demonstrating equivalent survival outcomes with primary radiation and primary surgery [8–10]. Even though patients receiving IOBT tended to have more advanced cT stage, the LRC, PFS, and OS were comparable between the two groups. This facet should give comfort to patients who often are eligible for either primary surgery or primary radiation approaches. Additionally, this study may help support aggressive salvage therapy for patients with oligometastatic disease in selected patients. One of three patients treated aggressively with recurrent, oligometastatic HPV+ had long term disease control. While preliminary, these findings parallel a recent report from Leeman et al. demonstrating that salvage therapy for oligometastatic head and neck cancer improved 2-year OS more than twofold [11]. Taken together, these data demonstrate that successful salvage of a solitary distant metastasis may lead to long term survival in a subset of patients. Interestingly, in our study as in the Leeman et al experience, patients with solitary metastatic disease—primarily to the lung—tend to represent the most favorable group. Further trials of patients with oligometastatic disease are warranted. Our study is limited by the inherent biases of a retrospective singleinstitution analysis, including the potential for selection biases, either in the patients that choose to come to our center for treatment, as well as in terms of the ultimate treatment patients receive, including things such as which patients are selected for aggressive treatment for oligometastatic disease. Despite these limitations, our data serves to validate the ≥5 lymph node cutoff selected in the AJCC 8th edition. It also helps to highlight that these patients are far more likely to fail distantly than locoregionally and supports the idea of aggressive treatment of patients with oligometastatic head and neck cancer.
Table 2 Cox proportional hazards univariate and multivariate analysis for progression free survival. Univariate analysis Variable
HR (95% CI)
Progression free survival Age: Continuous
1.02 (0.98–1.08)
Sex Male Female
1 [Ref.] 0.76 (0.17–3.35)
Location Tonsil Base of Tongue
1 [Ref.] 1.60 (0.63–4.07)
Smoker Never Ever
1 [Ref.] 1.30 (0.50–3.37)
Smoking dose ≤10py, Light > 10py, Heavy
1 [Ref.] 0.72 (0.27–1.93)
T stage cT1-2 cT3-4
1 [Ref.] 1.69 (0.60–4.80)
N stage (7th Ed. AJCC) pN0-2a pN2b-3
1 [Ref.] 1.60 (0.56–4.55)
Surgical margins − +
0.30
0.51
0.32
0.38
0.45 1 [Ref.] 2.33 (0.26–21.07)
1 [Ref.] 1.00 (0.38–2.65)
pNtotal: continuous Largest LN size: continuous
P
0.59
pENE − +
No. pN+ ≤4 ≥5
HR (95% CI)
0.32
1 [Ref.] 0.55 (0.20–1.47)
Adjuvant RT CRT
P
0.72
Primary treatment Brachytherapy Resection
Adjuvant Any None
Multivariate analysis
0.23
0.99
0.63 1 [Ref.] 1.64 (0.22–12.38) 0.26 1 [Ref.] 0.57 (0.21–1.52) 0.001 1 [Ref.] 5.46 (2.07–14.42) 1.03 (1.01–1.05) 1.05 (0.81–1.36)
0.005 0.69
0.005 1 [Ref.] 4.34 (1.57–12.02) 1.04 (1.00–1.05)
0.034
For the MVA, p = 0.20 was used as a cutoff and backwards exclusion was used. HR = Hazard Ratio. CI = Confidence Interval. PY = pack-year. pENE = pathologic extranodal extension. cT = clinical T stage. pN = pathologic N stage. AJCC = American Joint Committee on Cancer. − = Negative. + = Positive. Adjuvant = Adjuvant Therapy. pN+ = Number of pathologically positive lymph nodes. pNtotal = Total number of dissected lymph nodes. LN = Lymph node.
4%) in high-metastatic node number patients, they were unable to demonstrate a statistically significant association with this finding [4]. Additionally, our absolute rate of local recurrence and distant metastases were higher. This may be due to patient selection, as the Sinha et al. study consisted entirely of transoral resection patients, which typically has an upper limit to tumor size, suggesting a more favorable patient population compared to our institution where the majority of cases were open, allowing for greater resection of large tumors. In addition to validating the findings of the Haughey et al. staging paper [3], our study provides context by delineating patterns of failure,
Conclusions The AJCC 8th edition pathologic staging system classifications using ≥5 positive node number appears to predict well for oncologic outcomes. Distant metastases were the most common site of failure, particularly in patients with 5 or more involved nodes, suggesting that strategies to reduce DM risk in this group of patients are warranted. 38
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
Table 3 Patient Demographics of Progressed Patients. Pt #
Sex
Site
cT
pN
pN+
PY
1° Tx
Adjuv
Time to failure (mo)
Initial failure site (s)a
Salvage
Status
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Female Male Male Male Male Male Male Male Female Male Male Male Male Male Male Male Male
Tonsil BoT Tonsil BoT Tonsil Tonsil BoT BoT Tonsil Tonsil BoT BoT BoT BoT Tonsil BoT BoT
2 3 2 2 1 3 2 1 3 3 2 2 2 2 2 2 4
2a 2c 2b 2c 1 2c 2b 2b 2b 2c 2c 2c 2c 2c 3 2a 1
1 2 5 4 1 5 8 1 2 9 5 9 2 14 4 1 1
0 9 7.5 40 10 15 30 0 30 4 0 0 0 0 0 1.6 30
IOBT IOBT IOBT IOBT Resection Resection IOBT IOBT IOBT IOBT IOBT IOBT Resection IOBT Resection Resection Resection
RT RT CRT RT RT None CRT RT CRT CRT CRT CRT RT CRT CRT CRT None
7.2 12 3.7 11.8 35.6 11.9 9.9 20.5 3.7 59.5 16.4 16 41.6 18.5 9.7 11.0 12.0
Left Tonsillar Fossa Base of Tongue and Lung Liver Bilateral Lungs Pulmonary Nodules, Mediastinal LNs, Brain Mets Local Mass and Bilateral Lungs Multiple Bone Mets Lung and Spine Left Level III LN & 3 Lung Nodulesa LUL Lung Left Lung Left FoM, Level II LN, Testicle, and Humeral Heada RLL Lung Bilateral Lungs Lung Parotida Left Supraglottis and Lunga
Y N N N N N N N N N Y N Y Y N Y N
D D D D AWD AWD D D D AWD NED AWD AWD AWD AWD D D
pT = Pathologic T Stage. pN = 7th Ed. AJCC Pathologic N Stage. PY = pack years. 1° Tx = Primary Treatment (either brachytherapy [IOBT] or Resection). Adjuv = Adjuvant Therapy. Y = Yes, salvage attempted. N = N, no salvage. BoT = Base of Tongue. RT = Radiotherapy. CRT = Chemoradiotherapy. FoM = Floor of Mouth. LN = lymph node. LUL = left upper lobe. RLL = Right lower lobe. D = Deceased. AWD = Alive With Disease. NED = No Evidence of Disease. a The first failure site is documented; multiples are documented if they were the discovered first and synchronously.
Conflicts of interest
Oral Oncol 2016;62:11–9. [4] Sinha P, Kallogjeri D, Gay H, Thorstad WL, Lewis JS, Chernock R, et al. High metastatic node number, not extracapsular spread or N-classification is a node-related prognosticator in transorally-resected, neck-dissected p16-positive oropharynx cancer. Oral Oncol 2015;51:514–20. [5] Fakhry C, Zevallos JP, Eisele DW. Imbalance between clinical and pathologic staging in the updated american joint commission cancer staging system for human papillomavirus-positive oropharyngeal cancer. J Clin Oncol 2017. JCO2017752063. [6] Harari PM, Harris J, Kies MS, Myers JN, Jordan RC, Gillison ML, et al. Postoperative chemoradiotherapy and cetuximab for high-risk squamous cell carcinoma of the head and neck: Radiation Therapy Oncology Group RTOG-0234. J Clin Oncol 2014;32:2486–95. [7] Antonia SJ, Villegas A, Daniel D, Vicente D, Murakami S, Hui R, et al. Durvalumab after chemoradiotherapy in stage III non-small-cell lung cancer. N Engl J Med 2017. [8] Kelly JR, Park HS, An Y, Contessa JN, Yarbrough WG, Burtness BA, et al. Comparison of survival outcomes among human papillomavirus-negative cT1-2 N1–2b patients with oropharyngeal squamous cell cancer treated with upfront surgery vs definitive chemoradiation therapy: an observational study. JAMA Oncol 2017. [9] Soo K-C, Tan E-H, Wee J, Lim D, Tai B-C, Khoo M-L, et al. Surgery and adjuvant radiotherapy vs concurrent chemoradiotherapy in stage III/IV nonmetastatic squamous cell head and neck cancer: a randomised comparison. Br J Cancer 2005;93:279–86. [10] Iyer NG, Tan DSW, Tan VKM, Wang W, Hwang J, Tan N-C, et al. Randomized trial comparing surgery and adjuvant radiotherapy versus concurrent chemoradiotherapy in patients with advanced, nonmetastatic squamous cell carcinoma of the head and neck: 10-year update and subset analysis. Cancer 2015;121:1599–607. [11] Leeman JE, Li J-G, Pei X, Venigalla P, Zumsteg ZS, Katsoulakis E, et al. Patterns of treatment failure and postrecurrence outcomes among patients with locally advanced head and neck squamous cell carcinoma after chemoradiotherapy using modern radiation techniques. JAMA Oncol 2017;3:1487–94.
The authors have no conflicts of interest to disclose. Nicholas C.J. Lee: none declared. Jacqueline R. Kelly: none declared. Henry S. Park: none declared. Yi An: none declared. Benjamin L. Judson: none declared. Barbara A. Burtness: none declared. Zain A. Husain: none declared. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.oraloncology.2018.08. 001. References [1] Ang KK, Harris J, Wheeler R, Weber R, Rosenthal DI, Nguyen-Tân PF, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 2010;363:24–35. [2] Lydiatt WM, Patel SG, O’Sullivan B, Brandwein MS, Ridge JA, Migliacci JC, et al. Head and Neck cancers-major changes in the American Joint Committee on cancer eighth edition cancer staging manual. CA Cancer J Clin 2017;67:122–37. [3] Haughey BH, Sinha P, Kallogjeri D, Goldberg RL, Lewis JS, Piccirillo JF, et al. Pathology-based staging for HPV-positive squamous carcinoma of the oropharynx.
39
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
Patterns of failure in high-metastatic node number human papillomaviruspositive oropharyngeal carcinoma
T
Nicholas C.J. Leea, Jacqueline R. Kellya, Henry S. Parka, Yi Ana, Benjamin L. Judsonb, ⁎ Barbara A. Burtnessc, Zain A. Husaina, a
Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, USA Section of Otolaryngology, Department of Surgery, Yale School of Medicine, New Haven, CT, USA c Section of Medical Oncology, Department of Internal Medicine, Yale School of Medicine, New Haven, CT, USA b
A R T I C LE I N FO
A B S T R A C T
Keywords: Head and neck cancer Oropharyngeal carcinoma Human papillomavirus Distant metastasis Radiotherapy Surgery Neoplasm Oropharynx Squamous cell cancer
Background: The 8th edition American Joint Committee on Cancer staging system for resected HPV-positive oropharynx carcinoma (HPV+ OPC) highlights high node number as a critical determinant of survival. We sought to characterize outcomes and patterns of failure in patients with high pathologically involved node number oropharynx cancer. Methods: We retrospectively identified 116 HPV+ OPC patients sequentially treated with neck dissection and either resection or intraoperative brachytherapy of the primary tumor between 2010 and 2016. External beam radiation was given based on the pathologic findings. Cox proportional hazards regression was used for multivariate analysis. Results: With a median follow-up of 27 months, the 3-year overall survival and progression free survival (PFS) were 89% and 81%, respectively. On multivariate analysis, ≥5 involved lymph nodes was significantly associated with worse PFS (hazard ratio 4.3, 95% confidence interval (CI) 1.5–12.0, P = 0.001). Rates of 3-year locoregional recurrence (LRR) in patients with ≤4 vs ≥5 were 6% and 22% (log-rank P = 0.12). Rates of 3-year distant metastases (DM) were 12% and 53% between ≤4 and ≥5 (log-rank P < 0.001). Conclusion: Our findings confirm that patients with 5 or more involved lymph nodes appear to have substantially worsened rates of disease recurrence. While these patients appear to be at high risk of both LRR and DM, the predominant mechanism of failure is distant, and the rate of DM in this group was over 50%. Dedicated clinical trials in this patient population are warranted with a focus on mitigating the high DM rate.
Introduction Human papillomavirus-positive (HPV+) oropharyngeal carcinoma (OPC) compromises over 70% of all newly diagnosed OPC cases and exhibits more favorable survival outcomes compared to HPV-negative disease. Survival rates with HPV+ disease are approximately 30% higher at 5 years than with HPV-negative disease [1], which has led to the design of different staging systems for both HPV+ and HPV-negative disease. Additionally, amongst HPV+ patients, a separate staging system has been adopted for clinically staged versus pathologically staged patients [2]. In the new pathologic staging system, the number of involved nodes, as opposed to laterality or size, is a key determinant of stage in resected cancers. In the study that formed the basis for the
American Joint Committee on Cancer (AJCC) 8th edition staging system, the presence of ≥5 pathologically positive nodes was associated with worsened 5-year overall survival (OS) 71% vs 84% [3]. Yet, the patterns of failure within this high-risk group were not specifically elucidated—an important aspect for developing therapeutic strategies to counteract the worsened survival rates. We sought to further examine outcomes within a cohort of patients treated surgically with neck dissections and either resection or intraoperative brachytherapy of the primary site to examine risk factors for recurrence, validate the ≥5 node cutoff, characterize the role of salvage therapy in patients with recurrence, and, most importantly, examine patterns of failure in high (≥5) node number versus low (≤4) node number patients.
Presented in part at the 2017 American Society of Therapeutic Radiation Oncology Conference (September 26, 2017; San Diego, CA) and 2018 American Radium Society Conference (May 8, 2018; Orlando, FL). ⁎ Corresponding author at: Department of Therapeutic Radiology, Yale School of Medicine, 35 Park Street, New Haven, CT 06511, USA. E-mail address: [email protected] (Z.A. Husain). https://doi.org/10.1016/j.oraloncology.2018.08.001 Received 12 April 2018; Received in revised form 18 July 2018; Accepted 1 August 2018 1368-8375/ © 2018 Elsevier Ltd. All rights reserved.
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
Methods
Table 1 Patient demographics.
Patient population HPV-associated OPC patients from 2010 to 2016 at Yale New Haven Hospital (YNHH) who underwent neck dissection were included in this retrospective study. Those with previous head and neck cancer or cancer treatment to the head and neck region were excluded. Yale University’s IRB approved the protocol and written consent was waived. Pathologic data, HPV status, and clinicodemographic features were extracted from patient records. HPV status was determined using p16 immunohistochemistry and/or HPV in situ hybridization (ISH). Study endpoints Progression free survival (PFS) was defined as time from primary surgery to progression or death from any cause. OS was defined as time from primary surgery to death of any cause. Distant metastases (DM) were defined as disease outside of the head and neck region and was confirmed by imaging and/or biopsy-proven disease consistent with the primary when possible. Locoregional recurrence (LRR) was defined as recurrence within the head and neck. Statistical analysis Statistical analysis was performed using STATA/IC software package (version 14.2; StataCorp, College Station, TX). Multivariable analysis (MVA) was conducted using backwards-exclusion with the Cox proportional hazards model. The inclusion cut-off was P < 0.20. Proportionality was confirmed using time-dependent covariates. Survival analysis was conducted using the Kaplan-Meier method and log-rank test. Results Patient and tumor characteristics One hundred and sixteen patients were included with a median follow-up of 27 months (range: 1 to 63 months). Patient and tumor characteristics are described in Table 1. The median patient age was 60, and 101 (87%) patients were male. 64 (55%) of patients had a tonsil primary while 52 (45%) had a base of tongue primary. Ninety-eight (85%) were cT1-2, and 83 (72%) were 7th Ed. AJCC pN1-2b. 11.2% had greater than 4 pathologically involved lymph nodes (range 0 to 14). The median number of lymph nodes dissected was 34 (range: 5–97). Fiftynine patients (51%) had pENE. Forty-one patients (35%) received intraoperative brachytherapy (IOBT) to the primary site without resection while 75 patients (65%) received primary tumor resection. Of the resected patients, 35 underwent transoral robotic surgery (TORS). There was a strong trend for patients undergoing primary brachytherapy to having more advanced T stages than patients undergoing primary surgery (Supp. Table A). Ten patients (24%) receiving IOBT were cT3-4 while eight patients (11%) receiving primary tumor resection were cT34 (p = 0.051). Patients undergoing IOBT were treated with palladium103 with a median dose of 25.74 Ci (range: 17.75–44.4 Ci). Of those receiving resection, 11 had positive margins (14%). Overall, 64 patients (55%) received post-operative chemoradiation (CRT), 38 (33%) received adjuvant RT alone, and 13 (11%) received no post-operative therapy. Within those receiving no post-operative therapy, eight were Stage I-II and pN0 on neck dissection. Of those receiving chemotherapy, 48 (74%) received cisplatin, six (9%) received carboplatin, four (6%) received a combination, and eight (12%) were not definitively specified due to treatment at an outside facility. Adjuvant RT was delivered with IMRT at a median dose of 58 Gy (range: 50–70 Gy) in 30 fractions (median; range: 20–38). Amongst the 63 patients with radiation fields that were available for review, 59 patients received radiation to the
Characteristic
Total (%)
≤4 pN+ (%)
≥5 pN+ (%)
P
Total Age: median [range], y
116 60 [36–87]
103 (88.8) 59 [36–87]
13 (11.2) 62 [48–81]
0.14
Sex Male Female
101 (87.1) 15 (12.9)
88 (85.4) 15 (14.6)
13 (100) 0 (0)
Disease Site Tonsil and Soft Palate Base of Tongue
64 (55.2) 52 (44.8)
58 (56.3) 45 (43.7)
6 (46.2) 7 (53.8)
Smoking Never smoker Former smoker Current smoker
45 (38.8) 40 (34.5) 31 (18.7)
39 (37.9) 37 (35.9) 27 (26.2)
6 (46.2) 3 (23.1) 4 (30.8)
Smoking dose, pysa ≤10, Light > 10, Heavy
62 (54.9) 51 (45.1)
53 (51.5) 47 (45.6)
9 (69.2) 4 (30.8)
cTb T1 T2 T3 T4
40 (34.5) 58 (50.0) 12 (10.3) 6 (5.2)
37 (35.9) 50 (48.5) 10 (9.7) 6 (5.8)
3 8 2 0
(23.1) (61.5) (15.4) (0)
pN (AJCC 7th)c N0 N1 N2a N2b N2c N3
16 (13.8) 10 (8.6) 24 (20.7) 49 (42.2) 12 (10.3) 5 (4.3)
16 (15.5) 10 (9.7) 24 (23.3) 42 (40.8) 6 (5.8) 5 (4.9)
0 0 0 7 6 0
(0) (0) (0) (53.8) (46.2) (0)
Primary Treatment Brachytherapy Resection
41 (35.3) 75 (64.7)
33 (32) 70 (68)
8 (61.5) 5 (38.5)
pENE Negative Positive
59 (50.9) 57 (49.1)
49 (47.6) 54 (52.4)
10 (76.9) 3 (23.1)
Surgical margins Negative Positive
62 (86.1) 11 (13.9)
57 (83.8) 11 (16.2)
5 (1 0 0) 0 (0)
Adjuvant Therapyd None Chemotherapy RT CRT
13 (11.2) 1 (0.9) 38 (32.8) 64 (55.2)
52 (50.5) 1 (1.0) 38 (36.9) 12 (11.7)
12 (92.3) 0 (0) 0 (0) 1 (7.7)
0.21
0.49
0.71
0.27
0.68
< 0.001
0.06
0.08
> 0.99
> 0.99
y = year. pys = pack-years. cT = clinical T stage. pN = pathologic N stage. AJCC = American Joint Committee on Cancer. pN+ = number of pathologically positive lymph nodes. a Incomplete smoking data. b Comparison of cT1-2 v cT3-4. c Comparison of pN0-2a v pN2b-3. d Comparison of any adjuvant therapy to none. P values determined by Fisher’s exact test, chi-square, or student t-test where appropriate.
tumor bed and bilateral neck, while 4 patients received radiation to the tumor bed and ipsilateral neck only. Survival outcomes For all patients, the 3-year PFS and OS were 81% and 89% respectively. Of the 103 patients with ≤4 nodes, the 3-year PFS was 86%, while for the 13 patients with ≥5 involved nodes the 3-year PFS was 47% (Fig. 1; log-rank P = 0.002). However, 3-year OS was not significantly different in patients with ≤4 nodes at 90% and ≥5 nodes at 81% (log-rank P = 0.26). We found no significant difference in 3-year OS between patients with pENE (91%) and those without pENE (86%) (log-rank p = 0.30) or 3-year PFS (pENE+ 83% and 77%, log-rank P = 0.89). Among patients with resection of the primary tumor, there was no difference in 3-year 36
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
Fig. 1. Kaplan-Meier of Progression Free Survival by metastatic node number.
Notably, patients with ≥5 positive lymph nodes had a significantly higher rate of 3-year DM than patients with ≤4 nodes (53% versus 12%, p < 0.001).
OS between patients with positive versus negative margins. Patients had similar PFS, OS, and locoregional control (LRC) with IOBT or surgical management of the primary tumor (3-year PFS 74% v 80% logrank p = 0.22; 3-year OS 87% v 91%, log-rank p = 0.64; 3-year LRC 89% v 94%, log-rank p = 0.33, respectively). Additionally, there was no difference in 3-year OS based on 7th Ed. AJCC nodal staging (pN0-2a 89% v pN2b-3 88% log-rank p = 0.75). On univariate analysis of predictors of recurrence, two variables met criteria for inclusion in the PFS MVA: number of nodes dissected (continuous, P = 0.005) and the number of positive nodes (≤4 vs. ≥5, P = 0.001). After taking into account relevant clinicodemographic variables in our MVA, both were significant. The number of nodes dissected had a HR 1.04 and 95% CI 1.001–1.05. The lymph node threshold of ≥5 positive nodes remained highly significant (P = 0.005, HR 4.3, 95% CI 1.6–12.0) as demonstrated in Table 2.
Salvage therapy Five patients underwent salvage therapy with curative intent (median follow-up: 47 months; range: 23–69 months). Of these, two were patients with LRR and three were patients who recurred with oligometastatic disease that were treated with metastectomy and/or stereotactic body radiotherapy. Of the five patients, one patient had no evidence of disease (NED) at last follow up. This patient was originally a Stage IVB cT2N3M0 never-smoker who underwent upfront IOBT and bilateral neck dissection followed by external beam radiotherapy with cisplatin. Approximately one year and four months after finishing CRT, he was found to have an 8 mm left lower lobe (LLL) nodule on surveillance imaging. He underwent a salvage LLL metastectomy with pathology confirming two foci of p16-positive, HPV ISH-positive squamous cell carcinoma (8 and 7 mm). He is currently four years out from completion of treatment and remains NED.
Patterns of recurrence In total, seventeen patients (15%) developed disease recurrence (Table 3). Recurrences were detected by CT in six cases (35%), PET in ten cases (59%), and chest x-ray in one case confirmed by CT (6%). Biopsies were obtained in fifteen cases (88%), and all were consistent with the primary tumor. Seven patients developed locoregional recurrence: one was isolated, five were synchronous with DM, and one had LRR followed by DM. Of these seven LRR, five were ipsilateral and two were contralateral (both out of field). Of the ipsilateral LRR, three were in field, one out of field, and one did not receive RT. In addition, of the two patients who recurred in the neck, one was contralateral (in level III) and the other patient recurred at the same level of the initial disease presentation. Patients with ≥5 positive lymph nodes had a higher rate of 3-year LRR compared to ≤4 nodes, although this did not reach criteria for significance (22% versus 6%, respectively, log-rank P = 0.12). Sixteen patients developed DM: 10 were isolated, 5 were synchronous with LRR, and one occurred after an initial LRR. Of these, two patients developed disseminated metastases involving organs including the brain and testicles. The median time to recurrence was 1 year (range: 113 days–5 years). Three of our 16 patients (19%) who developed distant metastasis did so after at least two years of surveillance.
Discussion The 8th edition AJCC system for pathologically staged HPV+ OPC has made pathologic node number a key determinant of stage. This is based on work demonstrating that the AJCC 7th edition pN staging did not accurately stratify patients by oncologic outcome by survival or hazard analysis, findings that were also shown in our dataset [3,4]. In the multi-institutional study that became the basis for the 8th edition staging system, ≤4 LNs and ≥5 LNs were associated with 5-year OS values of 89% and 71%, respectively [3,5]. Here, the 3-year OS values were 90% and 81% for ≤4 LNs and ≥5 LNs, respectively, while the 3 year PFS values were 86% and 47% for ≤4 LNs and ≥5 LNs, respectively. The significant difference in PFS but not OS for our cohort is likely due to shorter follow-up compared to Haughey et al, which when combined with the pattern of late recurrences, likely did not allow for recurrences to convert into deaths [3]. A notable difference between our series and the Sinha et al. series is that while their experience also showed a four-fold increase DM (16% v 37
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
which helps inform therapeutic strategies for this high risk group of patients. Our work demonstrates that patients with 5 or more involved nodes are at risk for both LRR and DM, with values of 22% and 53% respectively. The most common site of failure for these patients is overwhelmingly distant. This finding is critical as it impacts the aim of clinical trials in this patient population. Currently, in ECOG 3311 (NCT01898494), all patients with 5 or more involved nodes receive concurrent cisplatin based XRT, with the idea being that these patients are at risk for LRR and adding radiosensitizing chemotherapy may be of benefit. This work suggests that in addition to a radiosensitizing agent for locoregional control, a focus on therapies that may affect the distant metastatic rate may be indicated. For example, concurrent docetaxel may be one such strategy. In RTOG 0234 (NCT00084318), a randomized phase II study, for high risk postoperative patients with HNSCC, compared cisplatin vs docetaxel with 60 Gy radiation [6]. This showed improved PFS in the docetaxel arm, largely due to a decrease in distant metastatic disease (2-year DM rate of 25% in cisplatin vs 13% in docetaxel). This strategy is being further investigated in the RTOG 1216 (NCT01810913) study, although it is aimed at a different population, namely high-risk (positive margin or pENE+), specifically excluding HPV+ oropharyngeal cancer patients. Alternatively, given the recent data suggesting a benefit to adjuvant immunotherapy in suppressing distant metastases in NSCLC after definitive local therapy, the ≥5 node group may represent a key demographic for adjuvant immunotherapy trials in HPV+ head and neck cancer [7]. Our data helps confirm other important findings in HPV+ OPC. Firstly, we demonstrate no significant difference in OS between patients treated with IOBT versus resection in HPV+ OPC, which is consistent with previous work in head and neck cancers demonstrating equivalent survival outcomes with primary radiation and primary surgery [8–10]. Even though patients receiving IOBT tended to have more advanced cT stage, the LRC, PFS, and OS were comparable between the two groups. This facet should give comfort to patients who often are eligible for either primary surgery or primary radiation approaches. Additionally, this study may help support aggressive salvage therapy for patients with oligometastatic disease in selected patients. One of three patients treated aggressively with recurrent, oligometastatic HPV+ had long term disease control. While preliminary, these findings parallel a recent report from Leeman et al. demonstrating that salvage therapy for oligometastatic head and neck cancer improved 2-year OS more than twofold [11]. Taken together, these data demonstrate that successful salvage of a solitary distant metastasis may lead to long term survival in a subset of patients. Interestingly, in our study as in the Leeman et al experience, patients with solitary metastatic disease—primarily to the lung—tend to represent the most favorable group. Further trials of patients with oligometastatic disease are warranted. Our study is limited by the inherent biases of a retrospective singleinstitution analysis, including the potential for selection biases, either in the patients that choose to come to our center for treatment, as well as in terms of the ultimate treatment patients receive, including things such as which patients are selected for aggressive treatment for oligometastatic disease. Despite these limitations, our data serves to validate the ≥5 lymph node cutoff selected in the AJCC 8th edition. It also helps to highlight that these patients are far more likely to fail distantly than locoregionally and supports the idea of aggressive treatment of patients with oligometastatic head and neck cancer.
Table 2 Cox proportional hazards univariate and multivariate analysis for progression free survival. Univariate analysis Variable
HR (95% CI)
Progression free survival Age: Continuous
1.02 (0.98–1.08)
Sex Male Female
1 [Ref.] 0.76 (0.17–3.35)
Location Tonsil Base of Tongue
1 [Ref.] 1.60 (0.63–4.07)
Smoker Never Ever
1 [Ref.] 1.30 (0.50–3.37)
Smoking dose ≤10py, Light > 10py, Heavy
1 [Ref.] 0.72 (0.27–1.93)
T stage cT1-2 cT3-4
1 [Ref.] 1.69 (0.60–4.80)
N stage (7th Ed. AJCC) pN0-2a pN2b-3
1 [Ref.] 1.60 (0.56–4.55)
Surgical margins − +
0.30
0.51
0.32
0.38
0.45 1 [Ref.] 2.33 (0.26–21.07)
1 [Ref.] 1.00 (0.38–2.65)
pNtotal: continuous Largest LN size: continuous
P
0.59
pENE − +
No. pN+ ≤4 ≥5
HR (95% CI)
0.32
1 [Ref.] 0.55 (0.20–1.47)
Adjuvant RT CRT
P
0.72
Primary treatment Brachytherapy Resection
Adjuvant Any None
Multivariate analysis
0.23
0.99
0.63 1 [Ref.] 1.64 (0.22–12.38) 0.26 1 [Ref.] 0.57 (0.21–1.52) 0.001 1 [Ref.] 5.46 (2.07–14.42) 1.03 (1.01–1.05) 1.05 (0.81–1.36)
0.005 0.69
0.005 1 [Ref.] 4.34 (1.57–12.02) 1.04 (1.00–1.05)
0.034
For the MVA, p = 0.20 was used as a cutoff and backwards exclusion was used. HR = Hazard Ratio. CI = Confidence Interval. PY = pack-year. pENE = pathologic extranodal extension. cT = clinical T stage. pN = pathologic N stage. AJCC = American Joint Committee on Cancer. − = Negative. + = Positive. Adjuvant = Adjuvant Therapy. pN+ = Number of pathologically positive lymph nodes. pNtotal = Total number of dissected lymph nodes. LN = Lymph node.
4%) in high-metastatic node number patients, they were unable to demonstrate a statistically significant association with this finding [4]. Additionally, our absolute rate of local recurrence and distant metastases were higher. This may be due to patient selection, as the Sinha et al. study consisted entirely of transoral resection patients, which typically has an upper limit to tumor size, suggesting a more favorable patient population compared to our institution where the majority of cases were open, allowing for greater resection of large tumors. In addition to validating the findings of the Haughey et al. staging paper [3], our study provides context by delineating patterns of failure,
Conclusions The AJCC 8th edition pathologic staging system classifications using ≥5 positive node number appears to predict well for oncologic outcomes. Distant metastases were the most common site of failure, particularly in patients with 5 or more involved nodes, suggesting that strategies to reduce DM risk in this group of patients are warranted. 38
Oral Oncology 85 (2018) 35–39
N.C.J. Lee et al.
Table 3 Patient Demographics of Progressed Patients. Pt #
Sex
Site
cT
pN
pN+
PY
1° Tx
Adjuv
Time to failure (mo)
Initial failure site (s)a
Salvage
Status
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Female Male Male Male Male Male Male Male Female Male Male Male Male Male Male Male Male
Tonsil BoT Tonsil BoT Tonsil Tonsil BoT BoT Tonsil Tonsil BoT BoT BoT BoT Tonsil BoT BoT
2 3 2 2 1 3 2 1 3 3 2 2 2 2 2 2 4
2a 2c 2b 2c 1 2c 2b 2b 2b 2c 2c 2c 2c 2c 3 2a 1
1 2 5 4 1 5 8 1 2 9 5 9 2 14 4 1 1
0 9 7.5 40 10 15 30 0 30 4 0 0 0 0 0 1.6 30
IOBT IOBT IOBT IOBT Resection Resection IOBT IOBT IOBT IOBT IOBT IOBT Resection IOBT Resection Resection Resection
RT RT CRT RT RT None CRT RT CRT CRT CRT CRT RT CRT CRT CRT None
7.2 12 3.7 11.8 35.6 11.9 9.9 20.5 3.7 59.5 16.4 16 41.6 18.5 9.7 11.0 12.0
Left Tonsillar Fossa Base of Tongue and Lung Liver Bilateral Lungs Pulmonary Nodules, Mediastinal LNs, Brain Mets Local Mass and Bilateral Lungs Multiple Bone Mets Lung and Spine Left Level III LN & 3 Lung Nodulesa LUL Lung Left Lung Left FoM, Level II LN, Testicle, and Humeral Heada RLL Lung Bilateral Lungs Lung Parotida Left Supraglottis and Lunga
Y N N N N N N N N N Y N Y Y N Y N
D D D D AWD AWD D D D AWD NED AWD AWD AWD AWD D D
pT = Pathologic T Stage. pN = 7th Ed. AJCC Pathologic N Stage. PY = pack years. 1° Tx = Primary Treatment (either brachytherapy [IOBT] or Resection). Adjuv = Adjuvant Therapy. Y = Yes, salvage attempted. N = N, no salvage. BoT = Base of Tongue. RT = Radiotherapy. CRT = Chemoradiotherapy. FoM = Floor of Mouth. LN = lymph node. LUL = left upper lobe. RLL = Right lower lobe. D = Deceased. AWD = Alive With Disease. NED = No Evidence of Disease. a The first failure site is documented; multiples are documented if they were the discovered first and synchronously.
Conflicts of interest
Oral Oncol 2016;62:11–9. [4] Sinha P, Kallogjeri D, Gay H, Thorstad WL, Lewis JS, Chernock R, et al. High metastatic node number, not extracapsular spread or N-classification is a node-related prognosticator in transorally-resected, neck-dissected p16-positive oropharynx cancer. Oral Oncol 2015;51:514–20. [5] Fakhry C, Zevallos JP, Eisele DW. Imbalance between clinical and pathologic staging in the updated american joint commission cancer staging system for human papillomavirus-positive oropharyngeal cancer. J Clin Oncol 2017. JCO2017752063. [6] Harari PM, Harris J, Kies MS, Myers JN, Jordan RC, Gillison ML, et al. Postoperative chemoradiotherapy and cetuximab for high-risk squamous cell carcinoma of the head and neck: Radiation Therapy Oncology Group RTOG-0234. J Clin Oncol 2014;32:2486–95. [7] Antonia SJ, Villegas A, Daniel D, Vicente D, Murakami S, Hui R, et al. Durvalumab after chemoradiotherapy in stage III non-small-cell lung cancer. N Engl J Med 2017. [8] Kelly JR, Park HS, An Y, Contessa JN, Yarbrough WG, Burtness BA, et al. Comparison of survival outcomes among human papillomavirus-negative cT1-2 N1–2b patients with oropharyngeal squamous cell cancer treated with upfront surgery vs definitive chemoradiation therapy: an observational study. JAMA Oncol 2017. [9] Soo K-C, Tan E-H, Wee J, Lim D, Tai B-C, Khoo M-L, et al. Surgery and adjuvant radiotherapy vs concurrent chemoradiotherapy in stage III/IV nonmetastatic squamous cell head and neck cancer: a randomised comparison. Br J Cancer 2005;93:279–86. [10] Iyer NG, Tan DSW, Tan VKM, Wang W, Hwang J, Tan N-C, et al. Randomized trial comparing surgery and adjuvant radiotherapy versus concurrent chemoradiotherapy in patients with advanced, nonmetastatic squamous cell carcinoma of the head and neck: 10-year update and subset analysis. Cancer 2015;121:1599–607. [11] Leeman JE, Li J-G, Pei X, Venigalla P, Zumsteg ZS, Katsoulakis E, et al. Patterns of treatment failure and postrecurrence outcomes among patients with locally advanced head and neck squamous cell carcinoma after chemoradiotherapy using modern radiation techniques. JAMA Oncol 2017;3:1487–94.
The authors have no conflicts of interest to disclose. Nicholas C.J. Lee: none declared. Jacqueline R. Kelly: none declared. Henry S. Park: none declared. Yi An: none declared. Benjamin L. Judson: none declared. Barbara A. Burtness: none declared. Zain A. Husain: none declared. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.oraloncology.2018.08. 001. References [1] Ang KK, Harris J, Wheeler R, Weber R, Rosenthal DI, Nguyen-Tân PF, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 2010;363:24–35. [2] Lydiatt WM, Patel SG, O’Sullivan B, Brandwein MS, Ridge JA, Migliacci JC, et al. Head and Neck cancers-major changes in the American Joint Committee on cancer eighth edition cancer staging manual. CA Cancer J Clin 2017;67:122–37. [3] Haughey BH, Sinha P, Kallogjeri D, Goldberg RL, Lewis JS, Piccirillo JF, et al. Pathology-based staging for HPV-positive squamous carcinoma of the oropharynx.
39