Second primary cancer after index head and neck squamous cell carcinoma in Northern China
Zhien Feng, MD, PhD, Qiao Shi Xu, MD, Li Zheng Qin, MD, PhD, Hua Li, MD, PhD, Xin Huang, MD, Ming Su, MD, and Zhengxue Han, MD, PhD
Objective. To evaluate the clinicopathologic features, prognostic factors, and management of patients in the North Chinese population with head and neck squamous cell carcinoma (HNSCC) who developed a second primary malignancy (SPM).
Methods. This was a retrospective study including 1818 eligible patients between June 1999 and April 2011.
Results. A total of 188 HNSCC patients developed SPM. Multiple oral dysplastic lesions (MODLs) (P < .001) were among the risk factors for occurrence of SPM. However, MODLs were closely associated with many mild pathologic features, such as early T stage (P < .001), early N stage (P ¼ .036), good pathologic differentiation (P < .001), and mild growth pattern (P < .001). Interestingly, multivariate survival analysis showed that SPM patients had a better prognosis if they had the characteristics of MODLs (P ¼ .020).
Conclusions. MODLs were a crucial risk factor leading to the occurrence of oral SPM after an index HNSCC in patients in Northern China. However, SPM patients with the characteristics of MODLs had a better prognosis. (Oral Surg Oral Med Oral Pathol Oral Radiol 2017;123:95-102)
Worldwide, approximately 635,000 new cases of head and neck cancer are diagnosed annually; more than 12%
of these cases occur in China. Unfortunately, more than 76,000 patients with head and neck cancer die each year, and the majority of these patients have head and neck squamous cell carcinoma (HNSCC).1 Today, second primary malignancy (SPM) is the leading cause of death for patients who experience long-term survival after an index HNSCC.2 Therefore, clarification of the incidence rate, risk factors, and specific mortality for SPM is crucial to further improve the prognosis of patients with HNSCC in China.
Recent data from Europe and the United States show that most SPMs after an index HNSCC tumor are located in the head and neck, lung, or esophagus.3 These epidemiologic data supported the concept of “field cancerization” as a reasonable interpretation of SPMs after HNSCC.4Recently, multiple oral dysplastic lesions (MODLs), especially the proliferative verrucous leukoplakia subtype, were also believed to be crucial factors for occurrence of SPM.5 However, it is well known that HNSCC is an obviously heterogeneous disease that varies across many characteristics, including
age, sex, ethnicity, region, diet, alcohol and tobacco use, tumor-node-metastasis (TNM) classification, histologic grade, treatment modality, prognosis, and so on.2,6-8 Currently there is little high-level information about epidemiologic data, risk factors, or prognosis for patients with HNSCC in Northern China, which has a population of more than 600 million people, or approximately half the Chinese population.9
The aims of this retrospective study were to inves- tigate the clinicopathologic features, prognostic factors, and management of SPM in the North Chinese popu- lation and describe our experience with this rare disease.
METHODS AND MATERIALS Study population
This study was approved by the Institutional Review Board of the Beijing Stomatological Hospital and was conducted in accordance with the World Medical Association’s Declaration of Helsinki (2002 version).
We retrospectively reviewed the clinicopathological data of patients with HNSCC who were treated in the Department of Oral and Maxillofacial Head and Neck Oncology, Beijing Stomatological Hospital of Capital Medical University, from June 1999 through April 2011. Patients who met the following criteria were
This work was supported by the National Natural Science Foundation of China (grants 81302350 and 81570957), the Beijing Science and Technology Committee (grant Z161100000516201), the Beijing Nova program (grant Z151100000315045), and the Discipline Con- struction Fund of Beijing Stomatological Hospital (grant 15-09-12).
Department of Oral and MaxillofacialeHead and Neck Oncology, Beijing Stomatological Hospital, Capital Medical University, Beijing, China.
Received for publication Jun 17, 2016; returned for revision Aug 4, 2016; accepted for publication Aug 11, 2016.
Ó 2016 Elsevier Inc. All rights reserved.
2212-4403/$ - see front matter
http://dx.doi.org/10.1016/j.oooo.2016.08.013
Statement of Clinical Relevance
This study highlights evaluation of prognostic fac- tors and management of second primary malignancy associated with head and neck squamous cell carcinoma.
95
included in the study: (1) they had all been affected by a previously untreated HNSCC; (2) they had all under- gone curative surgical therapy; and (3) the sites of the primary index HNSCC included the tongue, lower gingiva, buccal mucosa, floor of the mouth, oropharynx, upper gingiva, and hard palate. Finally, a total of 1818 eligible patients with HNSCC were included in this study. The cases were restaged ac- cording to the 2002 version of the Union for Interna- tional Cancer Control/American Joint Committee on Cancer classification based on the initial clinical description and computed tomography (CT), magnetic resonance imaging, chest X-ray,flexible esophagogas- troduodenoscopy, or positron emission tomography examination.
Definition of SPM, MODLs, and diffuse infiltration SPM wasfirst defined by Warren and Gates in 1932.10 The latest criteria for SPM were modified by the National Cancer Institute as follows: SPM is defined as a metachronous, invasive, solid cancer developing
6 months after an index HNSCC.11 Specifically, if the second cancer originated from a nonsquamous cell or developed in a different location, or if the SPM developed in the same region more than 5 years after the index cancer diagnosis, it is coded as SPM.
MODLs are defined as multifocal dysplastic lesions that occur on the oral mucosa and are known to be potentially malignant, including multifocal leukoplakia (e.g., proliferative verrucous leukoplakia), eryth- roplakia, erythroleukoplakia, and submucousfibrosis of the oral mucosa.12,13
Diffuse infiltration, a subtype of histologic signs of severity as described in our previous study, is defined as diffusely invasive growth of primary tumor cell observed under a microscope.1
Treatment and pathologic analysis
All patients were initially treated with surgery. The protocols for surgery, radiotherapy (RT), and concomi- tant chemoradiotherapy were consistent with our previ- ous description.7For routine histopathologic analysis of primary tumors and neck dissection specimens, all primary tumors and each node section were placed in different groups and subjected to standard hematoxylin and eosin staining.14
Follow-up strategy
The patients were regularly followed up as described previously.15 The patients underwent semiannual chest X-ray and other imaging examinations (ultrasonography, CT, magnetic resonance imaging, positron emission tomography/CT, and/or flexible
esophagogastroduodenoscopy). If a recurrent or second primary or newly developed malignant lesion was suspected, other diagnostic modalities were utilized to confirm these lesions.
Statistical analysis
The follow-up study continued until April 1, 2016.
Patients who were lost to follow-up within 1 year of surgery were excluded from the statistical analysis. The baseline demographic data between comparable subgroups were compared using the chi-square test for categorical variables and the t test for continuous variables. The primary outcome assessment parameter was 5-year disease-specific survival (DSS). Statistical significance was tested using the log-rank test. Uni- variate and multivariate analyses were used to identify the independent predictors of SPM and 5-year DSS.
Independent prognosticators were identified by multi- variate Cox regression analysis using the forward se- lection method. All tests were two-sided, and P< .05 was considered statistically significant. Statistical soft- ware (SPSS, version 17.0; Chicago, IL, USA) was used for all statistical analyses.
RESULTS
Patient characteristics
A total of 1818 cases of primary HNSCC between June 1999 and April 2011 were identified by retrospective retrieval from the database. The patient and tumor characteristics are summarized inTable I. The median follow-up time for patients with HNSCC was 66 months, and 277 patients (15.2%) were lost to follow-up. At the end of follow-up, 188 of the 1818 patients (10.3%) developed SPM, 1354 patients (74.5%) did not develop SPM, and the SPM status of 276 patients (15.2%) was unknown because of loss to follow-up. The median time until occurrence of SPM was 49 months after thefirst operation. Specifically, the cumulative incidence of SPM was 27.7% within 2 years, 31.4% at 2 to 5 years, and 41.0% beyond 5 years after the index tumor surgery. Of the seven primary subsites in the study, the hard palate (ratio of SPM/non-SPM¼ 6.4%/2.9%; 2.2) and buccal mucosa (ratio of SPM/non-SPM ¼ 21.8%/14.3%; 1.5) had a relatively higher probability of developing post- operative SPM. Furthermore, SPM locations after index HNSCC were the oral cavity (n ¼ 127, 67.6%), head and neck other than oral cavity (n ¼ 7, 3.7%), lung (n¼ 9, 4.8%), esophagus (n ¼ 17, 9.0%), breast (n ¼ 5, 2.7%), uterus (n¼ 3, 1.6%), liver (n ¼ 5, 2.7%), col- orectum (n¼ 5, 2.7%), penis (n ¼ 2, 1.1%), stomach (n ¼ 4, 2.1%), and one each (0.5%) in the leg, gallbladder, kidney, and bladder. Notably, 32 of these patients experienced three to six primary cancers after
Table I. Baseline data of patients and second primary malignancies (SPMs) in the study
Variable
Total (n¼ 1818) Non-SPM (n¼ 1354) SPM (n¼ 188)*
P value
No. (%) No. % No. %
Age, years 59 (15-89) 58.8 12.2 58.7 11.3 .912
Gender
Male 1077 (59.2) 809 59.7 98 52.1 .047
Female 741 (40.8) 545 40.3 90 47.9
Site
Tongue 675 (37.1) 534 39.4 56 29.8 .002
Lower gingiva 294 (16.2) 216 16.0 33 17.6
Buccal mucosa 284 (15.6) 194 14.3 41 21.8
Floor of the mouth 191 (10.5) 136 10.0 22 11.7
Oropharynx 140 (7.7) 101 7.5 14 7.4
Upper gingiva 174 (9.6) 134 9.9 10 5.3
Hard palate 60 (3.3) 39 2.9 12 6.4
T stage
T1 439 (24.1) 328 24.2 61 32.4 .080
T2 687 (37.8) 512 37.8 72 38.3
T3 201 (11.1) 144 10.6 18 9.6
T4a 466 (25.6) 349 25.8 36 19.1
T4b 25 (1.4) 21 1.6 1 0.6
Pathologic Nodal status
N0 839 (46.1) 623 55.2 93 64.1 .004
N1 305 (16.8) 230 20.4 35 24.1
N2 371 (20.4) 274 24.3 17 11.8
N3 2 (0.1) 2 0.1 0 0.0
Nx (No ND)* 301 (6.6) - - - -
Pathologic grade
I 887 (48.8) 674 51.3 101 54.9 .194
II 777 (42.7) 565 43.0 78 42.4
III 104 (5.7) 76 5.7 5 2.7
Missing 50 (2.8) - - - -
Growth pattern
Exophytic 583 (32.0) 406 33.9 77 43.8 .033
Ulcerative 560 (30.8) 399 33.3 53 30.1
Infiltrative 505 (27.8) 392 32.8 46 26.1
Missing 170 (9.4) - - - -
Smoking history
Smoker 777 (42.7) 588 45.1 70 38.0 .073
Nonsmoker 980 (53.9) 717 54.9 114 62.0
Missing 61 (3.4) - - - -
Alcohol history
Drinker 578 (31.8) 436 33.4 54 29.3 .272
Nondrinker 1179 (64.9) 869 66.6 130 70.7
Missing 61 (3.4) - - - -
Extracapsular spread
Absence 301 (73.8) 255 75.0 27 84.4 .236
Presence 107 (26.2) 85 25.0 5 15.6
Perineural invasion
Absence 587 (82.3) 458 81.8 71 88.8 .124
Presence 126 (17.7) 102 18.2 9 11.2
Vascular/lymphatic emboli
Absence 697 (98.3) 546 98.0 80 100.0 .375
Presence 12 (1.7) 11 2.0 0 0.0
Diffuse infiltration
Absence 530 (74.8) 416 74.7 61 76.3 .763
Presence 179 (25.2) 141 25.3 19 23.7
MODLs
Absence 1614 (88.8) 1247 92.1 132 70.2 <.001
Presence 182 (10.0) 107 7.9 56 29.8
Missing 22 (1.2) - - - -
SPM, second primary malignancy; SD, standard deviation; MODL, multiple oral dysplastic lesion.
*The number of known SPM patients was 188.
an unfortunate SPM. Ultimately, of 188 patients who experienced SPM, 124 patients received surgical treat- ment, 27 patients received surgery plus adjuvant radiotherapy, 23 patients received palliative radio- therapy and/or chemotherapy, and the remaining 14 patients terminated therapy.
Multiple oral dysplastic lesions are a risk factor leading to SPM occurrence
By the chi-square test, close correlations were seen be- tween the cumulative incidence of SPM and the following clinicopathological parameters: sex (P¼ .047), primary site (P¼ .002), pathologic nodal status (P ¼ .004), growth pattern (P¼ .033), and MODLs (P < .001). However, there were no correlations between SPM and tobacco use (P¼ .073) or alcohol use (P ¼ .272) in the study. Based on logistic regression analysis (forward method), the values of the associated factors (i.e., sex, primary site, pathologic nodal status, growth pattern, and MODLs) in predicting the development of SPM were further evalu- ated. MODLs (positive correlation, odds ratio: 4.563, 95% CI: 2.793-7.453, P < .001) and pathologic nodal status (negative correlation, odds ratio: 0.723, 95% CI:
0.563-0.930, P ¼ .012) constituted risk factors for the occurrence of SPM.
MODLs are closely associated with more mild clinicopathologic features of HNSCC
Considering that MODLs were the only positive cor- relation factor resulting in SPM, we further analyzed the associations between MODLs and other clinico- pathologic factors in 1796 of the 1818 patients whose MODL data were available. There were significant as- sociations when occurrences of MODL were compared with regard to tumor and patient characteristics, including female sex (P < .001), buccal sites (P ¼ .012), early T stage (P < .001), early N stage (P¼ .036), good pathologic differentiation (P < .001), mild growth pattern (P < .001), less tobacco use (P< .001), less alcohol use (P < .001), and less diffuse infiltration (P ¼ .025), reflecting the heterogeneity of HNSCC across MODLs (Table II).
A total of 56 patients who developed SPM had pre- vious MODL disease. In this subgroup, 31 patients (55.4%) had exophytic type (papillary architecture); 11 patients (19.6%) had ulcerative type; 13 patients (23.2%) had infiltrative type; and data were missing for 1 patient (1.8%). The SPM locations for patients with MODLs after an index HNSCC were the tongue (n¼ 12, 21.4%), lower gingiva (n ¼ 12, 21.4%), buccal mucosa (n ¼ 14, 25.0%), floor of the mouth (n ¼ 5, 8.9%), oropharynx (n¼ 3, 5.4%), upper gingiva (n ¼ 4, 7.1%), and hard palate (n ¼ 6, 10.8%). Twenty-three
patients developed multiple primary carcinomas, including third carcinoma (n ¼ 17, 73.9%), fourth carcinoma (n¼ 4, 17.4%), and fifth carcinoma (n ¼ 2, 8.7%). For treatment after the first recurrence, 19 pa- tients received surgery alone and the remaining 4 pa- tients received surgery plus adjuvant radiotherapy.
Survival analysis
During the follow-up period, 953 (52.4%) of the 1818 patients survived, 589 patients (32.4%) died, and 276 patients (15.2%) were lost to follow-up. Forty-seven patients died due to causes unrelated to cancer, including 22 patients who died of cardiac failure or cerebral stroke, 9 patients who died of multiple organ failure, 10 patients who died of respiratory failure, 2 patients who died of acute gastrointestinal haemor- rhage, 1 patient who died of suicide, 2 patients who died of uncertain causes, and 1 patient who died of septicemia.
In the entire cohort, the 5-year DSS rate was 58.9%.
Generally, there was no significant difference when comparing SPM patients with non-SPM patients in DSS (53.7% vs 66.4%, P ¼ .356, Figure 1). However, Kaplan-Meier analysis showed that patients whose survival time was longer than 5 years experienced a sharp decrease in DSS rate if they developed SPM (from 97.0% to 68.8%, P < .001, Figure 1).
Furthermore, patients with SPM located in the head and neck region compared to other sites had better DSS (head and neck SPM vs nonehead and neck SPM: 63.4% vs 29.6%, P< .001, Figure 2).
As mentioned above, MODLs were a predisposing factor for the development of SPM, and HNSCC pa- tients showed obvious heterogeneity of MODLs. The association between MODL status and prognosis was further analyzed. Interestingly, Kaplan-Meier analysis found that HNSCC patients with previous MODLs (74.8%) had better DSS than those without MODLs (63.7%) (P< .001,Figure 3).
HNSCC patients with SPM had a better prognosis if they had the characteristics of MODLs
To evaluate high-risk factors for a poor prognosis for SPM patients, the baseline data served as covariates and were analyzed using Cox proportional hazards regres- sion models. A univariate analysis of the 188 SPM patients showed that sex (P¼ .001), T stage (P < .001), pathologic nodal status (P ¼ .001), pathologic grade (P ¼ .002), tobacco use (P < .001), alcohol use (P< .001), and no history of MODLs (P ¼ .032) were high-risk prognostic factors for determining the DSS of SPM patients. A further multivariate survival analysis showed that alcohol use (hazard ratio [HR]: 2.874, 95%
CI: 1.743-4.737, P< .001), high pathologic grade (HR:
1.755, 95% CI: 1.270-2.426, P¼ .001), and no history of MODLs (HR: 0.430, 95% CI: 0.211-0.876, P ¼ .020) were independent prognostic factors for worse DSS in SPM patients. That is, HNSCC patients with SPM had a better prognosis if they had the char- acteristics of MODLs (Table III).
Prognostic scoring of risk factors and screening of high-risk populations
Prognostic scoring of risk factors for DSS included alcohol use, high pathologic grade, and no history of MODLs. In this study, each risk factor for SPM identi- fied as an independent prognosticator in survival analysis (alcohol use, high pathologic grade, and no history of
MODLs) was given a score of 1. The DSS rate differed significantly between patients with a score of 0 (61.4%) and a score of 1 (47.7%) or 2 (42.9%). Therefore, patients with a score of 1 or2 were identified as the high-risk population for DSS, and those with a score of 0 were identified as the low-risk population.
Surgery-based salvage treatment for resectable SPM patients could result in good outcomes By further analysis between prognosis and different treatments in the low-risk population (score of 0), we found that patients with SPM who underwent surgery alone had a similar DSS to those who underwent surgeryþ RT (69.4% vs 77.8%, P ¼ .678,Figure 4). In the high-risk population (score of 1 or2), the patients Table II. Associations between multiple oral dysplastic lesions (MODLs) and clinicopathologic factors in the study (MODLs known [n = 1796]*)
Variable
Patients without MODLs (n¼ 1614) Patients with MODLs (n¼ 182)
P value
No. % No. %
Sex
Male 986 61.1 75 41.2 <.001
Female 628 38.9 107 58.8
Site
Tongue 592 36.7 75 41.2 .012
Lower gingiva 268 16.6 23 12.6
Buccal mucosa 138 14.7 41 22.5
Floor of the mouth 177 10.0 11 6.0
Oropharynx 130 8.1 8 4.4
Upper gingiva 157 9.7 16 8.8
Hard palate 52 3.2 8 4.4
T stage
T1 353 21.9 83 45.6 <.001
T2 618 38.3 66 36.3
T3 182 11.3 14 7.7
T4a 438 27.1 18 9.9
T4b 23 1.4 1 0.5
Pathologic Nodal status
N0 756 54.3 70 68.6 .036
N1 284 20.4 16 15.7
N2 351 25.2 16 15.7
N3 2 0.1 0 0.0
Pathologic grade
I 741 47.3 137 76.1 <.001
II 730 46.6 40 22.2
III 97 6.1 3 1.7
Growth pattern
Exophytic 484 33.4 93 52.2 <.001
Ulcerative 505 34.9 46 25.8
Infiltrative 459 31.7 39 21.9
Smoking history
Smoker 722 46.4 41 22.9 <.001
Nonsmoker 834 53.6 138 77.1
Alcohol history
Drinker 543 34.9 28 15.6 <.001
Nondrinker 1013 65.1 151 84.4
Diffuse infiltration
Absence 480 73.6 48 87.3 .025
Presence 172 26.4 7 12.7
*Of all 1818 patients, the number of cases with multiple oral dysplastic lesions was 1796.
who underwent surgery alone also had a DSS similar to those who underwent surgeryþ RT (59.7% vs 55.6%, P ¼ .647, Figure 4). In contrast, patients who underwent only palliative radiotherapy/chemotherapy or who stopped therapy had a less favorable DSS regardless of whether they were in the low-risk or high-risk population, but especially if they were in the high-risk population (low-risk vs high-risk: 23.5% vs 0.0%, marginal difference found, P¼ .087, Figure 4).
The results showed that surgical salvage treatment for resectable SPM patients could result in a good outcome, regardless of high-risk or low-risk score and whether they accepted adjuvant radiotherapy.
DISCUSSION
Although diagnosis and treatment techniques have greatly improved in the past three decades, the prog- nosis for HNSCC patients remains poor.1 SPM has become the chief culprit that threatens long-term sur- vival.16The objective of this study was to evaluate the clinicopathologic features, prognostic factors, and management of patients in the North Chinese Fig. 1. The survival curve between second primary malig-
nancy (SPM) and non-SPM groups for head and neck squa- mous cell carcinoma (HNSCC) patients.
Fig. 2. The survival curve of second primary malignancy (SPM) patients between cases in the head and neck region and those not in the head and neck region.
Fig. 3. The survival curve of head and neck squamous cell carcinoma (HNSCC) patients according to presence or absence of previous multiple oral dysplastic lesions (MODLs).
Table III. High-risk factors for disease-specific survival (DSS) among head and neck squamous cell carcinoma (HNSCC) patients who experienced second primary malignancy (SPM)
Variable Hazard ratio
95% Confidence interval P value Univariate analysis
Sex (male vs female) 0.471 0.303-0.732 .001 T stage (T1, T2, T3, T4a,
T4b)
1.453 1.196-1.765 <.001 pN (N0, N1, N2, N3) 1.373 1.131-1.667 .001 Pathologic grade (I, II, III) 1.680 1.218-2.318 .002 Tobacco habit (absence vs
presence)
2.296 1.487-3.546 <.001 Alcohol habit (absence vs
presence)
2.774 1.789-4.302 <.001 Multiple oral dysplastic
lesions (MODLs) (absence vs presence)
0.578 0.350-0.954 .032
Multivariate survival analysis Alcohol habit (absence vs
presence)
2.874 1.743-4.737 <.001 Pathologic grade (I, II, III) 1.755 1.270-2.426 .001 MODLs (absence vs
presence)
0.430 0.211-0.876 .020
population with HNSCC who developed SPM. The results of our study showed that patients whose survival time was longer than 5 years experienced a sharp decrease in DSS rate if they developed SPM, and the occurrence of SPM was closely associated with MODLs. Interestingly, the presence of MODLs was associated with a better prognosis for SPM patients, and only surgery-based salvage treatment for resectable SPM patients could result in good outcomes.
In this study, the occurrence of SPM was closely associated with MODLs, consistent with the theory of field cancerization. Because betel quid chewing is rare in North China, the pathogenesis, clinicopathologic features, and prognosis of HNSCC are significantly different from those of patients in Southern Asia.17 Currently, the tobacco/alcohol-related field cancer- ization theory has been widely accepted as explaining the incidence of SPMs in nonebetel quid-chewing areas.13 A study by Hamadah et al. showed that approximately 3-24% of patients with multiple primary malignancy had a medical history of oral precancerous lesions.13,18 Tobacco and alcohol habits have been widely identified as major etiologic factors in the populations of Western and Southern Asian coun- tries.19-21 However, these habits were not closely correlated with SPM in the Northern Chinese popula- tion, perhaps because half of the SPM patients in the study were female and had never been exposed to tobacco or alcohol.
Our results showed that alcohol habit, pathologic grade, and MODLs were factors closely correlated with the prognosis of SPM patients. Of these three factors, the presence of MODLs was the sole positive factor
predicting better survival, which could be explained by MODLs being closely associated with mild clinico- pathologic features in HNSCC patients, such as early T or N stage, good pathologic differentiation, mild growth pattern, less diffuse infiltration, and so on. Additionally, this importance was attributed to the more superficial nature of the lesions in patients with MODLs, stricter follow-up frequency, and stronger cancer-prevention awareness.12 Therefore, the presence of MODLs was associated with a better prognosis for SPM patients.
These results were consistent with Akrish’s conclusions, which demonstrated that MODLs presented with significantly better prognostic factors and short-term survival rates and longer duration of disease.22
Some recent studies have shown that the burden of SPM is high in patients with HNSCC, with more than 168 second solid tumors developing per 10,000 person- years at risk,2and a comparison between the prognosis of SPMs located in the head and neck region and those not in the head and neck region found that the former had better outcomes than the latter.16 Those results were similar to our results. However, our study found that the head and neck region, especially the oral cavity, was the most common subsite involved in SPM. The results were different from the study of Birkeland et al.,16 who reported that almost three- quarters of SPMs are located outside the head and neck region. However, survival analysis has demon- strated that patients whose survival time was longer than 5 years experienced a sharp decrease in DSS rate if they developed SPM. The results further indicated that SPM was the leading cause of death for patients with HNSCC who experienced long-term survival.2
Based on analysis of different risk populations, surgery-based salvage treatment for resectable SPM patients could also result in a good prognosis. This conclusion was the same as the viewpoint of Strojan et al.: “Whenever feasible, salvage surgery is the method of choice for curative intent; patients at high risk for local recurrence should be advised that post- operative radiotherapy or re-radiotherapy is expected to increase locoregional control at the expense of higher toxicity and without survival advantage compared to salvage surgery alone.”23 Therefore, we recommend that HNSCC patients who experience SPM should adopt an aggressive management strategy.
This study was retrospective and thus had inherent limitations. It could also be criticized for a lack of data about some important baseline factors, including severity of tobacco and alcohol exposure, depth of in- vasion, and tumor thickness. These limitations will be further considered in future studies. However, because large sample sizes have been analyzed and different variables evaluated, some generalizations are possible.
Fig. 4. The survival curves of low-risk and high-risk patients who developed second primary malignancy (SPM) and received different management (surgery alone vs surgeryþ radiotherapy vs palliative or terminated therapy).
CONCLUSIONS
MODLs were a crucial risk factor leading to the occurrence of SPM after an index HNSCC in patients in Northern China. However, SPM patients with MODLs had a better prognosis. For HNSCC patients who develop SPM, an aggressive surgery-based strategy of management should be recommended if the SPM is resectable.
REFERENCES
1.Feng Z, Guo W, Zhang C, et al. CCND1 as a predictive biomarker of neoadjuvant chemotherapy in patients with locally advanced head and neck squamous cell carcinoma. PLoS One. 2011;6:
e26399.
2.Morris LG, Sikora AG, Patel SG, Hayes RB, Ganly I. Second primary cancers after an index head and neck cancer: Subsite- specific trends in the era of human papillomavirus-associated oropharyngeal cancer. J Clin Oncol. 2011;29:739-746.
3.Leon X, Pedemonte G, Garcia J, Lopez M, Martel M, Quer M.
Elective treatment of the neck for second primary tumors of the head and neck. Eur Arch Otorhinolaryngol. 2014;271:1187-1190.
4.Ha PK, Califano JA. The molecular biology of mucosal field cancerization of the head and neck. Crit Rev Oral Biol Med.
2003;14:363-369.
5.Cerero-Lapiedra R, Balade-Martinez D, Moreno-Lopez LA, Esparza-Gomez G, Bagan JV. Proliferative verrucous leukopla- kia: A proposal for diagnostic criteria. Med Oral Patol Oral Cir Bucal. 2010;15:e839-e845.
6.Hashibe M, Brennan P, Chuang SC, et al. Interaction between tobacco and alcohol use and the risk of head and neck cancer:
Pooled analysis in the International Head and Neck Cancer Epidemiology Consortium. Cancer Epidemiol Biomarkers Prev.
2009;18:541-550.
7.Feng Z, Niu LX, Yuan Y, Peng X, Guo CB. Risk factors and treatment of contralateral neck recurrence for unilateral oral squamous cell carcinoma: A retrospective study of 1482 cases.
Oral Oncol. 2014;50:1081-1088.
8.Hawk NN, Long TE, Imam MH, et al. Clinicopathologic features and outcome of young adults with stage IV colorectal cancer. Am J Clin Oncol. 2015;38:543-549.
9.Niu LX, Feng Z, Li JN, Li CZ, Peng X, Guo CB. Prognostic factors of squamous cell carcinoma of the buccal mucosa: A retrospective study of 168 cases in North China. J Oral Max- illofac Surg. 2014;72:2344-2350.
10.Warren S, Gates O. Multiple primary malignant tumors: A survey of the literature and a statistical study. Am J Cancer. 1932;16:
1358-1414.
11.Curtis RE, Ries LA. Methods. In: Curtis RE, Freedman DM, Ron E, et al., eds. New Malignancies Among Cancer Survivors:
SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute; 2006:9-14.
12.Feng Z, Xu QS, Niu QF, et al. Risk factors for patients with multiple synchronous primary cancers involving oral and
oropharyngeal subsites. Oral Surg Oral Med Oral Pathol Oral Radiol. 2016;121:360-366.
13.Hamadah O, Goodson ML, Thomson PJ. Clinicopathological behaviour of multiple oral dysplastic lesions compared with that of single lesions. Br J Oral Maxillofac Surg. 2010;48:503-506.
14.Feng Z, Qin LZ, Huang X, Li JZ, Su M, Han Z. Nodal yield: Is it a prognostic factor for head and neck squamous cell carcinoma?
J Oral Maxillofac Surg. 2015;73:1851-1859.
15.Feng Z, Li JN, Li CZ, Guo CB. Elective neck dissection versus observation in the management of early tongue carcinoma with clinically node-negative neck: A retrospective study of 229 cases.
J Craniomaxillofac Surg. 2014;42:806-810.
16.Birkeland AC, Rosko AJ, Chinn SB, Prince ME, Sun GH, Spector ME. Prevalence and outcomes of head and neck versus non-head and neck second primary malignancies in head and neck squamous cell carcinoma: An analysis of the surveillance, epidemiology, and end results database. ORL J Otorhinolaryngol Relat Spec. 2016;78:61-69.
17.Liao CT, Wang HM, Ng SH, et al. Good tumor control and survivals of squamous cell carcinoma of buccal mucosa treated with radical surgery with or without neck dissection in Taiwan.
Oral Oncol. 2006;42:800-809.
18.Thomson PJ, Hamadah O. Cancerisation within the oral cavity:
The use of’field mapping biopsies’ in clinical management. Oral Oncol. 2007;43:20-26.
19.Thomas G, Hashibe M, Jacob BJ, et al. Risk factors for multiple oral premalignant lesions. Int J Cancer. 2003;107:285-291.
20.Dietrich T, Reichart PA, Scheifele C. Clinical risk factors of oral leukoplakia in a representative sample of the US population. Oral Oncol. 2004;40:158-163.
21.Warnakulasuriya KA, Robinson D, Evans H. Multiple primary tumours following head and neck cancer in southern England during 1961-98. J Oral Pathol Med. 2003;32:443-449.
22.Akrish S, Ben-Izhak O, Sabo E, Rachmiel A. Oral squamous cell carcinoma associated with proliferative verrucous leukoplakia compared with conventional squamous cell carcinoma: a clinical, histologic, and immunohistochemical study. Oral Surg Oral Med Oral Pathol Oral Radiol. 2015;119:318-325.
23.Strojan P, Corry J, Eisbruch A, et al. Recurrent and second pri- mary squamous cell carcinoma of the head and neck: When and how to reirradiate. Head Neck. 2015;37:134-150.
Reprint requests:
Zhengxue Han, MD, PhD
Department of Oral and MaxillofacialeHead and Neck Oncology Beijing Stomatological Hospital
Capital Medical University No. 4 Tian Tan Xi Li Dongcheng District Beijing 100050 P.R. China
Hanf1989@hotmail.com
