Introduction
Head and neck cancer (HNC) is the seventh most common cancer worldwide. It involves all malignancies originating from the anatomic subsites of the upper aerodigestive tract, including the oral cavity, larynx, oropharynx and hypopharynx [
1]. Globally, there were 4,100,000 prevalent cases and 1,100,000 new cases of HNC in 2016, leading to approximately 500,000 deaths [
2]. Of these malignancies, SCC constitutes 90% of histological types in HNC [
3]. The median age of patients is approximately 60 years, yet, an increasing trend of young-onset HNSCC in people younger than 45 years was reported worldwide. This may be due to the increased incidence of young-onset tumors affecting the oral cavity and oropharynx [
4‐
6].
The cancer treatment guidelines are not significantly differentiated between young and older patients. Surgery remains the mainstay of treatment for young-onset HNSCC, emphasizing performing en-bloc resection with a clear pathological margin greater than 5 mm, and achieving a good aesthetic and functional outcome and quality of life [
7,
8]. After surgical resection, adjuvant radiotherapy with standard dose and volume recommended by guidelines [
9] serves as a critical supplementary treatment for subclinical foci elimination. It is reserved for patients with adverse features, such as advanced T stage, multiple positive lymph nodes, lymphovascular or perineural space invasion, and positive margins [
10]. However, it has been reported in many cancers that PORT could result in elevated risks of long-term adverse consequences, including SPM [
11‐
13].
It should be noted that SPM represents the leading long-term cause of death among HNSCC patients, attributing to approximately 30% of overall deaths, triple the number of mortalities resulting from distant metastases [
14,
15]. Only a few studies have explored the association between radiotherapy and SPM incidence in HNSCC. Nevertheless, the results were equivocal. Some studies have demonstrated an increased risk of field cancerization due to radiotherapy [
16], but others suggested that radiotherapy could reduce the incidence of SPM in HNSCC [
17,
18]. Thus, no conclusive evidence has been found on the association between radiotherapy and SPM incidence in HNSCC. This may be caused by the following limitations: short follow-up, small sample size, or failure to consider a 5-year minimum latency span between radiation exposure and SPM development.
The adolescent and young population are particularly vulnerable to radiation-associated impacts due to their increased tissue susceptibilities and longer life expectancies [
19,
20]. Therefore, the impact of postoperative radiation on SPM development requires further investigation among these patients. This study aimed to examine the risk of SPM related to postoperative radiation for young-onset HNSCC using the US SEER-9 database and incorporate approximately more than 15 years of the follow-up period.
Discussion
PORT is an effective treatment to improve local control and prolong survival [
10]. Nevertheless, adolescent and young patients are especially vulnerable to radiation-induced carcinogenesis and possess longer life expectancies to develop radiation-associated SPM [
23]. This SEER-based study investigated the impact of PORT on developing SPM among adolescent and young patients who survived more than 5 years after the diagnosis of index HNSCC. We found a significantly positive correlation between the receipt of PORT and the long-term risk of developing SPM. Additionally, we conducted an internal comparison for the risk of developing SPM according to the receipt of PORT and further supported that the receipt of PORT was associated with an elevated risk of SPM, particularly exposure at a younger age, and the risk increased with longer follow-up. This study underscores the importance of long-term surveillance for these patients.
Evaluation of SIR and EAR, which was the highlight of most previous population-based studies on SPM [
14,
29], could provide meaningful preliminary population-level data to support investigations on the risk of developing SPM relative to the United States general population. In this SEER-based study using cancer surveillance data from 1975 to 2011, we observed a relatively elevated excess risk of developing SPM among PORT-treated patients, with an EAR of 107.99 compared to that of 69.73 among non-PORT-treated patients. Accordingly, the highest risk ratio of SPM among PORT-treated patients was observed in the head and neck (SIR, 23.95), followed by esophagus (SIR, 9.82), liver (SIR, 4.52), and lung and bronchus (SIR, 3.25).
However, the disparities between early-onset HNSCC patients and the general population, including tobacco use, socioeconomic status, and health care access, may bias and confound the results of external comparisons [
30‐
32]. Thus, the second objective of our investigations focused on the internal comparison, which was the greatest strength of this study compared with previous studies [
14,
29]. Notably, since PORT was not randomly assigned in this retrospective study, the adverse cancer features that differed in PORT-treated vs non-PORT-treated patients (e.g. pT4 primary tumor, pN2 or pN3 nodal disease, and positive margin) were likely to confound the internal cohort comparisons. Nevertheless, the internal comparison was less likely to be confounded by the tobacco use and drinking history as these factors did not impact the receipt of PORT for index HNSCC [
10]. Apart from the adverse cancer features, physician preference may be the major factor influencing the choice of PORT for HNSCC [
1], and it is not likely to impact the risk of developing SPM.
Through internal cohort comparisons, elevated PORT-associated risks were observed for the cancers of head and neck (RR, 1.19), esophagus (RR, 2.91), lung and bronchus (RR, 1.67), and thyroid (RR, 2.48), which were closed to irradiation fields and were more likely to be exposed to radiation leakage [
33]. The development of techniques from three-dimensional conformal radiotherapy (3D-CRT) to intensity-modulated radiotherapy (IMRT) has allowed more accurate dose distributions, thus minimizing treatment toxicity and improving the survival rates [
34]. However, more irradiation fields and monitor units were required for IMRT to modulate fluency, leading to longer irradiation times and increased dose for collimator scatter and head leakage [
33]. The impact of IMRT on the risk of developing SPM requires further investigation. Previous studies of other cohorts exposed to radiation have reported a significant dose–response between the exposure of radiation and SPM [
35,
36]. This study showed a reduced RR among patients diagnosed after 2000, indicating shorter follow-up periods and lower levels of radiation leakage from the improved radiation techniques.
The overall radiation-related risks of developing SPM differed by index HNSCC subsites, which were elevated among cases with index oral cavity SCC or laryngeal SCC, but not significantly decreased among patients with index oropharyngeal SCC and hypopharyngeal SCC. Specifically, for the patients with index oral cavity SCC, the highest burden of SPM was recognized in the region of head and neck (SIR, 48.34; RR, 1.88). For patients diagnosed with index laryngeal SCC or hypopharyngeal SCC, strongly elevated burdens of SPM were observed in esophagus (LSCC [SIR, 4.49; RR, 1965.58]; HPSCC [SIR, 55.6; RR, 144.88]), lung and bronchus (LSCC [SIR, 5.01; RR, 1]; HPSCC [SIR, 18.73; RR, 458.03]). These differences in the index HNSCC subsites were consistent with previous studies [
14,
37,
38]. The variations observed may be caused by the exposure of adjacent tissues to a higher dose of radiation leakage.
Compared with other races, the black race was related to a higher risk of developing SPM (HR, 2.21), which was in agreement with previous studies [
39]. However, black patients with index young-onset HNSCC were less likely to suffer from radiation-induced SPM, which required confirmation by further large-scale clinical observation. Regarding sex, male patients were more prone to developing SPM than female patients due to a higher prevalence of tobacco smoking among male patients, similar to previous reports [
31]. A higher radiation-related risk of SPM was also observed among male patients, warranting further confirmations and investigations.
RRs and HRs stratified by the follow-up times and the cumulative incidence curve consistently showed increased long-term risks (elevating with time) for radiation-associated SPM, which were in agreement with previous studies [
35,
36]. A higher radiation-associated risk of developing SPM was observed among index HNSCC patients with younger age (age ≤ 35 years; RR 1.44) and with localized diseases (RR, 1.16), as these patients had a longer life expectancy to be affected by the radiation exposure. This finding was consistent with our result that the radiation-associated risk of SPM was increased with time.
Multiple studies have reported the negative influence of SPM on the survival of patients with index HNSCC [
40‐
42]. Indeed, our study shows that the development of SPM leads to a lower survival outcome in both the PORT- and non-PORT-treated patients. Furthermore, we reveal that PORT increases the long-term risk of SPM. The findings of our study can help improve the follow-up surveillance strategies, which will benefit the adolescent and young patients with HNSCC, especially those with oral cavity SCC or laryngeal SCC. Notably, these findings should be interpreted cautiously, and further investigations and assessments of PORT in managing young-onset HNSCC are warranted to verify our results.
Several important limitations of this study should be considered. First, similar to most databases, the SEER had intrinsic selection bias and unmeasured confounding. Since PORT is administrable in outpatient clinics, PORT would likely be under ascertained if not documented, which may skew the estimations of RR toward the null. Furthermore, lack of the detailed data on postoperative radiation (e.g. the dose of radiation) has limited our investigation of the dose–response relationship between PORT and the development of SPM, although PORT is performed under the guidance of cancer treatment guidelines in most cases, which have recommended standard dose and volume. Future prospective studies should investigate the risk adapted de-intensification of postoperative radiotherapy for young patients in an effort to maximize survival benefits. Information on family history of cancer, smoking history, alcohol consumption, and health care access are also lacking, which could provide additional insights. Future studies are warranted to evaluate the correlation between these factors and PORT-associated risks of SPM development. Finally, previous studies into field tumorigenesis have reported abnormalities (e.g. loss of heterozygosity, high Ki-67 proliferation index, and TP53 gene mutation) in histologically normal tissues adjacent to the cancer of head and neck [
43]. It is possibly that cancer recurred within the area of preexisting genetic field tumorigenesis. Thus, our study focused on patients who survived longer than 5 years after diagnosing with index young-onset HNSCC. Despite these limitations, this is the first population-based study to assess the PORT-related SPM risk among 5 year survivors of index young-onset HNSCC. The methodological strengths of our study, including the large size of samples, near-complete follow-up duration, and the internal comparisons, maximize the generalizability and validity of results.
Taken together, this study shows that the development of SPM leads to a lower survival outcome in adolescent and young patients with index HNSCC. Furthermore, PORT is revealed to increase the long-term risk of SPM, based on our study of 5 year survivors with index young-onset HNSCC.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.