Abstract
Background/Aim: This study aimed to evaluate the outcomes of patients with head and neck squamous cell carcinoma (HNSCC) who underwent resection and refused the recommended adjuvant therapy. Patients and Methods: Locoregional recurrence-free survival (LRRFS) and time to progression (TTP) were assessed in HNSCC patients treated with surgery who declined some or all adjuvant therapy (refusal group (RG)) compared to those who received the recommended adjuvant therapy (TG). Results: With a median follow-up of 23 months, the 2-year LRRFS was significantly lower in the 17 patients from the RG compared to the 152 patients from the TG: 23.1% vs. 69%, HR=0.30, 95% confidence incidence (CI)=0.15-0.59; p<0.001. The mean TTP was 12 months in the RG and was not reached in the TG (p<0.001). Conclusion: Patients with HNSCC who declined the recommended adjuvant therapy had a recurrence rate of 50% within a year.
Adjuvant radiation therapy is the standard of care for resected head and neck squamous cell carcinoma (HNSCC) with adverse risk factors, such as the presence of multiple involved nodes, inadequate surgical margins, pT3-4 disease, lymphovascular invasion or perineural invasion due to improved local control (1-4). Two randomized trials have supported adjuvant chemoradiation in patients with extranodal extension or positive margins due to an overall survival benefit (5, 6). Despite these potential benefits, studies demonstrate that not all patients receive guideline-based adjuvant therapy (7). The reasons for this are not well known. A study identified some factors for patients not receiving the indicated adjuvant treatment for breast cancer including age above 70 years, comorbidity, mistrust of medical institutions, lack of adjuvant treatment benefits and concern about toxicities (8). Other likely factors are logistical concerns regarding costs and daily rides to treatment, as well as an inadequate understanding of the true risk of recurrence following curative-intent surgery, when all gross disease has been extirpated. There are limited data regarding outcomes in patients who refuse adjuvant therapy, particularly regarding the rate of recurrence, patterns of recurrence, salvage rates, and time to recurrence. We, therefore, sought to characterize outcomes for patients with resected head and neck squamous cell carcinoma who partially or completely declined the recommended adjuvant treatment (radiation or chemotherapy) to inform patient and physician decision making.
Patients and Methods
Patient population. This was a retrospective study approved by the Institutional Review Board (IRB) of the Yale School of Medicine. We included consecutive patients with HNSCC of the oral cavity, oropharynx, hypopharynx, and larynx who underwent surgery at Yale New-Haven Hospital/Smilow Cancer Center between May 2013 and July 2017 and were recommended to have adjuvant radiation or chemoradiation.
Patients who received adjuvant therapy as prescribed, made up the treated per recommendation group and information regarding radiation volume, doses, start and end of treatment were noted. Patients who refused adjuvant radiation and those who declined adjuvant chemotherapy while accepting adjuvant radiation were included in the refusal group. Patients who discontinued radiation after receiving less than half of the planned dose were also included in the refusal group. The reason(s) for refusal were retrieved either from the medical chart, the physician notes, or on occasion from the social work notes. Exclusion criteria included prior head and neck radiation, receipt of neoadjuvant chemotherapy, and patients with distant metastases.
Statistical analyses. Locoregional recurrence-free survival (LRFS), distant recurrence-free survival (DRFS) and progression-free survival (PFS) were defined as the time from the end of surgery until local, locoregional, distant or any progression and/or death respectively.
Overall survival (OS) was defined as the time between the date of surgery and the date of death or for living patients, the date of last follow-up. Patients who did not experience any events were censored at the date of last follow-up. Time- to-event endpoints were estimated using Kaplan–Meier methods. Statistical differences between cohorts were tested by the two-tailed log-rank test. In order to assess the accuracy of the obtained estimates, hazard ratios and 95% confidence intervals (CIs) were assessed from the Cox proportional hazards models. Cox analyses were used to identify significant predictors of locoregional progression, overall progression and death by assessing the following variables: age (≤65 vs. >65), sex, p16, extracapsular nodal extension (ENE), Union for International Cancer Control (UICC) stage (I/II vs. III/IV) 7th edition, margins status (negative vs. close/positive), lymphovascular invasion (LVI), perineural invasion (PNI) and refusal of adjuvant therapy. Factors with p<0.20 on univariate analysis were included in the multivariate analysis.
The treated per recommendation group and the refusal group were compared using the Chi-squared test and Fisher's exact test. T test or Wilcoxon rank sum test were performed to compare the parametric and the non-parametric quantitative parameters, respectively. Statistical significance was defined as p<0.05. All analyses were performed using R version 3.3(R-project, Vienna, Austria).
Results
Population. Between May 2013 and July 2017, 221 patients underwent curative-intent resection for primary treatment of invasive HNSCC at Yale-New Haven Hospital. Postoperative radiation therapy indications existed in 184 patients. Of these, 169 patients had available follow-up data and were otherwise eligible for inclusion. Amongst this group, 152 received treatment at the Yale-New Haven Hospital as prescribed and were part of the treated per recommendation group. In this group, 61 patients (44.8%) received chemoradiation. Seventeen patients declined the recommended adjuvant treatment and were included in the refusal group. Fourteen declined radiation and three declined chemotherapy while accepting radiation. Among the 14 patients who declined radiation, 11 did not receive any radiation dose and 3 discontinued after receiving less than half of the planned dose (12 Gy, 19 Gy and 30 Gy) citing side-effects as the reason for discontinuation.
Baseline characteristics in the two groups.
Indications for adjuvant therapy in the refusal group.
The characteristics of the two groups are shown in the Table I. No significant differences (p>0.05) were observed between the two groups in terms of classic locoregional prognostic factors: age, gender, smoking history, primary tumor, tumor stage, p16 status, margins status, PNI, LVI and ENE. The recommendation criteria for adjuvant therapy in the RG are presented in Table II. There were eleven patients with risk features in the RG: 10 patients with ENE and one with positive margins.
Reasons for declining therapy. In the refusal group, 15 patients declined adjuvant therapy because of concerns regarding potential side-effects, or early experience with side-effects in the case off the three patients who discontinued treatment early. The concern about potential side effects was so strong that three patients declined to even consult with a radiation oncologist. The two remaining patients declined adjuvant therapy due to concerns regarding costs of treatment.
Outcomes. With a median follow-up of 23 months, the 2-year LRRFS was significantly lower in the refusal group compared to the treated per recommendation group: 23.1% vs. 69%, HR=0.30, (95%CI=0.15-0.59) p<0.001. The median LRRFS was 14 months in the refusal group and was not reached in the treated per recommendation group.
The 2-year DRFS was decreased in the refusal group compared to the treated per recommendation group: 58% vs. 76.3%, HR=0.5 (95%CI=0.23-0.99), p=0.04. The median DRFS was 25 months in the refusal group and not reached in the treated per recommendation group.
The 2-year PFS was significantly lower in the refusal group compared to treated per recommendation group: 30% vs. 74.6%, HR=0.35 (95%CI=0.18-0.69), p=0.001. The median PFS was 14 months in the refusal group and not reached in the treated per recommendation group (Figure 1).
The absolute overall recurrence rate was significantly higher in the refusal group compared to the treated per recommendation group: 58.8 % (10/17) vs. 16.4% (25/152) (p<0.001). Mean TTP was 12 months in the refusal group and was not met in the treated per recommendation treatment group (p<0.001). Among the eight p16+ patients in the refusal group, three developed recurrence (37.5%).
Progression-free survival in the treated per recommendation and the refusal groups.
The 2-year OS was not statistically different in the refusal group and the treated per recommendation group: 74.1 % vs. 81%, HR=0.7 (95%CI=0.27-1.8), p=0.5. The median overall survival was 45 months in the refusal group and not reached in the treated per recommendation group. Five patients (29%) in the refusal group died from disease progression. Twenty-seven patients (17.8%) in the treated per recommendation group died.
Predictive factors of locoregional progression, overall progression and death. The results of the univariate and multivariate analyses for predictors of locoregional progression, overall progression and death are presented in Table III. In the multivariate Cox analyses, treatment refusal and p16 negativity were significantly associated with increased risk of locoregional progression with HR=4.84 (95%CI=2.1-12.4), p<0.001 and HR=5.84 (95%CI=2.07-16.3), p<0.001 respectively. Likewise, refusal and p16 negativity were associated with increased risk of overall progression with HR=3.40 (95%CI=1.61-7.17), p=0.001 and HR=3.56 (95%CI=1.53-8.3), p=0.003, respectively. However, only p16 negativity and ENE were found to be significantly associated with increased risk of death with HR=3.69 (95%CI=1.45-9.3), p=0.006 and HR=2.84 (95%CI=1.21-6.66), p=0.02, respectively.
Univariate and multivariate analysis for locoregional progression, overall progression and death.
Salvage therapy. Salvage curative-intent treatment was proposed in four out of 10 locoregionally recurrent cases in the refusal group. All these four patients were p16 negative. Two were treated with chemoradiation, one with radiation alone, and one case required surgery followed by chemoradiation. The median follow-up for these four patients was 9.5 months after salvage therapy and all patients were free of disease at the time of their last follow-up.
Discussion
Our results provide critical information regarding both the outcomes of patients who forgo recommended adjuvant therapy after undergoing curative-intent resection of their head and neck cancer. Critically, they demonstrate that patients who forgo recommended adjuvant therapies are at a high-risk of recurrence-nearly 60% in this study, with a mean time to recurrence of less than one year.
There are only a few publications evaluating the impact of the refusal of radiation and chemotherapy in the adjuvant setting and these mostly deal with cancers outside of the head and neck. A previous study investigated the impact of refusal of radiation therapy in several cancer types (mainly breast but also head and neck, lung, prostate and skin tumors) and reported compromised survival in refusal patients (9). Additionally, a recent study reported decreased survival in patients with breast, lung, colorectal or prostate cancer who refuse conventional cancer treatment in favor of alternative medicine (10). The same team found that patients who received complementary medicine were more likely to refuse the recommended treatment (11). Fear of potential complications represents one of the main reasons for refusal. The literature shows that patients not receiving the recommended treatment were less likely to know of the potential survival benefits (8). In our study, 15 patients declined adjuvant therapy because of the potential side effects and of these three even refused to meet a radiation oncologist. Two patients did not undergo adjuvant therapy because of the financial burden, which is also a recognized factor for non-compliance (12). In addition to patients who completely decline adjuvant treatment, non-compliance during treatment is also problematic. A retrospective study from a tertiary center reported a low adherence to planned adjuvant radiation treatment in a series of 81 patients. The theoretical effective dose received was less than 65 Gy in 48/81 patients (59%) with eight patients (10%) receiving less than 50 Gy. Only 33 patients (48%) received an effective dose greater than 65 Gy (13).
Specifically in the context of head and neck cancer, a study drawn from the National Cancer Database demonstrated increased mortality in patients not receiving adjuvant radiation in HNSCC (7). The limitation with these data however, is the lack of information regarding disease control and factors associated with non-use of post-operative therapy, as omission of standard of care post-operative therapy may have been due to comorbidities which constituted a competing cause of mortality.
Our results demonstrate that omission of adjuvant treatment due to patient choice is associated with significantly poorer disease control. Refusal was a strong independent predictor of locoregional progression with a hazard ration of 5.7. The locoregional progression and overall progression rate exceeded 50% in the observed cohort with a mean time to progression of 12 months, suggesting that most patients who decline adjuvant treatment will recur, and often within one year. This indicates that even for patients with comorbid conditions and a limited life expectancy, adjuvant therapy may be of benefit. Moreover, of the five deaths in the refusal group, all occurred in the setting of disease recurrence, suggesting that the PFS difference is being driven by head and neck cancer, not comorbid conditions.
Of the 17 patients in the refusal group, salvage local therapy was only recommended in four out of 17 (24%) patients, suggesting that deferral of initial therapy in favor of close follow up with salvage therapy reserved for disease recurrence is an unreliable strategy applicable to only a small number of patients. That said, all four patients (p16-negative) who underwent salvage surgery remain without evidence of disease, albeit with a short follow-up of only 9.5 months, suggesting that salvage should be considered in patients who are clinically eligible. These results are concordant with other recent studies demonstrating that salvage surgery in the modern era seems to hold more promise than in previous eras, particularly in HPV+ patients (14).
Given the recent increase in HPV+ OPSCC and the favorable prognosis associated with this disease, whether traditional risk factors for adjuvant therapy still apply, and whether adjuvant radiation can be withheld in some patients is an area of intense interest. A recent multi-institutional retrospective study reviewed the outcomes of 53 HPV+ oropharyngeal squamous cell cancer patients with traditional indications for adjuvant therapy, but who did not receive it, mainly (66%) due to patient refusal. Among all 53 patients, the 3-year cumulative incidence of relapse was 26.0%. The 3-year cumulative incidence was 11.8% in the 37 intermediate-risk patients (perineural invasion, lymphovascular invasion, T3-T4, or N2 disease) and 52.4% in the 16 high-risk patients (ENE or positive margins) (15). This argues that the traditional high-risk features are still valid for recurrence even in P16+ oropharyngeal cancer patients. Further study will be needed regarding the risk of forgoing radiation in the intermediate risk group.
Our study has some important limitations. Firstly, this is a retrospective study, and thus is subject to selection biases. There was an imbalance between numbers of patients in the two cohorts with a small sample size in the refusal group, although this is likely reflective of the clinical reality. The follow-up is modest but sufficient to find a meaningful difference in locoregional control and PFS. With longer follow-up, a survival benefit may have been detected as has been observed in another prior study (7). Another limitation was the fact that we included in the refusal group patients who started radiation but stopped their course prematurely because the given dose was unlikely to be therapeutic. The reason for stopping radiation in all three patients was related to side effects.
There are limited data exploring the clinical impact of refusing adjuvant treatment in head and neck cancer. The results observed in this study are significant and fill a gap in knowledge which can help patient and physician decision making.
Conclusion
In our study, refusal of recommended adjuvant radiation and/or chemotherapy resulted in significantly worse outcomes, with a recurrence rate of over 50% at the 1 year mean time point. These data may better inform HNSCC patients and their physicians in regard to the risks of declining recommended adjuvant therapy.
Acknowledgements
The Authors would like to thank their patients.
Footnotes
Authors' Contributions
GL and ZH designed the study. GL wrote the main manuscript text and prepared the Figure. GL, BK and WGY were involved with data collection. GL and ZH performed the data analysis. HSP, BLJ, YB, WGY, BB and Z.H. provided project supervision. All Authors have approved the final article.
This article is freely accessible online.
Conflicts of Interest
The Authors have no conflicts of interest to declare regarding this study.
- Received June 29, 2019.
- Revision received July 9, 2019.
- Accepted July 10, 2019.
- Copyright© 2019, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
In this issue
Jump to section
Related Articles
Cited By...
- Accelerated Fractionation With Concomitant Boost vs. Conventional Radio-chemotherapy for Definitive Treatment of Locally Advanced Squamous Cell Carcinoma of the Head-and-Neck (SCCHN)
- Comparison of Conventional Fractionation and Accelerated Fractionation With Concomitant Boost for Radiotherapy of Non-metastatic Stage IV Head-and-Neck Cancer