This article is mentioned in:

Introduction

Head and neck cancer affects ~686,000 new patients annually, 375,000 of whom eventually succumb to the disease according to the latest data from the World Health Organization (1). These data indicate that head and neck squamous cell carcinoma (HNSCC) is the sixth most common type of cancer worldwide. Thus, a growing number of patients diagnosed with head and neck cancer are confronted with not only the risk of local and regional relapse following curative therapy, but also the occurrence of a synchronous or metachronous secondary carcinoma. According to GLOBOCAN, the incidence of secondary carcinoma in different localizations in the head and neck is as follows: Oral cavity 4.3, larynx 2.2, pharynx 2.0, and nasopharynx 1.2 per 100,000 (2,3). Mortality depends on the localization of the primary tumor and the stage of the disease. Globally, a total of 4.58% of all cancer patients succumb to head and neck carcinoma (2).

Multiple primary malignancies is a term used to describe the occurrence of two or more malignancies at different anatomical localizations. These are defined as synchronous malignancies if the secondary carcinoma is diagnosed within 6 months of the primary carcinoma, or as metachronous if it is diagnosed >6 months after the primary carcinoma (4). In a retrospective study of 6,545 oncological patients in China, a secondary malignancy was identified in 72 cases (1.1%). In the present study, the highest incidence of secondary malignancies was noted among patients with head and neck cancer, followed by colorectal carcinoma and lung tumors (5). A retrospective study from India analyzed the data of 23,260 patients and found that 41 patients (0.18%) had developed a secondary carcinoma; these secondary malignancies were most common among patients with head and neck cancer, followed by gynecological tumors (6). An analysis from the USA, including 2,116,163 patients, identified 170,865 (8.1%) cases with a secondary carcinoma; in that study, the most common localizations of a secondary malignancy were the lung (18%), colon (12%), prostate (9%), and bladder (8%) (7). These differences confirm a wide variation in different parts of the world. Precise epidemiological data regarding the incidence and distribution of synchronous and metachronous malignancies are lacking, and the incidence in the current literature varies between 1.8 and 34.6% (8-19). A study from the Department of Health and Human Services in USA analyzed 58,363 patients with head and neck cancer between 1973 and 2008, and revealed that 11.7% (n=6,855) of the patients developed a secondary tumor. Patients with laryngeal carcinoma had the highest incidence of secondary tumors outside the head and neck area (13.6%), whereas patients with primary oral cancer had the highest incidence of secondary malignancies within the head and neck area [oral cavity (58.4%), larynx (18.2%) and oropharynx (13.3%)] (8). Metachronous carcinomas developed more often compared with synchronous carcinomas and, according to the literature, their frequency varied between 4.7 and 24%, whilst synchronous tumors were found in 0.3-14% of the patients (9,11,13,15,20). Di Martino et al categorized secondary tumors based on the time of diagnosis. Their retrospective study investigated 1,803 patients with primary tumors in the head and neck area, among whom 34 (1.9%) were found to have a synchronous carcinoma and 86 (4.8%) developed a metachronous carcinoma. Of those 86 patients, 50 developed the metachronous tumor within 5 years after primary tumor diagnosis, 22 patients developed the metachronous tumor within 6-10 years, and 14 patients after >10 years. Whether metachronous carcinomas with such long latency should be classified as secondary malignancies, with an etiological and pathogenic connection to the primary carcinoma, or as independent primary tumors, warrants further discussion (13).

The most important risk factors for HNSCC are alcohol and nicotine abuse, the chewing of betel nuts, insufficient dental hygiene, and infection with human papillomavirus, particularly the high-risk subtypes 16 and 18 (21-26). Smoking and drinking are major contributors to field cancerization and the occurrence of secondary malignancies (27-29). According to Ruback et al the combination of alcohol and tobacco abuse increases the risk of SCC up to 40-fold compared with either alcohol or tobacco alone. It may be inferred that alcohol serves as a dissolvent for some tobacco carcinogens and facilitates cellular uptake (30). A study by Pezzuto et al investigated the combined use of alcohol and tobacco in 11,221 patients with head and neck cancer; 72% of the patients claimed to have consumed tobacco or alcohol regularly; 4% of those patients claimed to have consumed alcohol alone, 33% tobacco alone, and 35% reported consuming both regularly (31). Both carcinogenic substances have a dose-effect relationship and cause cumulative toxicity. Carcinogenic effects may appear for even up to 10 years after the cessation of drinking and smoking (27). Smokeless forms of tobacco, such as snuffed or chewed tobacco, sometimes combined with betel quid or other preparations, such as the areca nut, are particularly consumed in India, North and South America and northern Europe. Betel nuts contain several alkaloids (guvacin, arecaidin, guvacoline and arecoline) that are addictive and have carcinogenic properties; they also inhibit tumor suppressor genes and DNA-repair mechanisms, causing uncontrolled cell proliferation within the mucosa and, thus, increasing the risk of cancer of the oral cavity and esophagus (28,32-35).

The development of secondary malignancies in the head and neck area may be explained by field cancerization. This term was coined in 1953 by Jaiswal et al, who observed multiple areas of SCC within the oropharyngeal cavity of a patient. Field cancerization describes the occurrence of multiple sites with invasive growth and areas of dysplasia within the mucosa. Despite the metacentric origin, field cancerization appears to be an important factor for the occurrence of secondary SCCs after therapy completion (21).

Secondary malignancies occur frequently in patients with head and neck carcinoma and may seriously affect the prognosis of the patients, which warrants a close follow-up; therefore, we herein sought to evaluate the frequency of secondary malignancies among head and neck cancer patients in our center and identify possible clinical indicators of their occurrence following curative tumor therapy. The aim of the present study was to help predict the occurrence of secondary cancer within smaller cohorts of other certified tumor centers.

Patients and methods

Ethics statement

The present study was conducted in full accordance with the Declaration of Helsinki and was approved by the Ethics Committee II of the Medical Faculty of Mannheim at the University of Heidelberg, Mannheim, Germany (file no. 2016-827R-MA).

Patients

A total of 380 patients with HNSCC who were treated at University Hospital Mannheim between 2006 and 2015 were enrolled. Patients for whom no complete follow-up data could be obtained, or those who died during the follow-up period, were excluded. Patients with secondary carcinomas other than SCC, i.e., sarcomas, lymphomas, etc., as well as patients receiving palliative treatment were also excluded. All included patients underwent a thorough clinical inspection, including a physical and endoscopic examination of the head and neck area, as well as sonographic and radiological imaging for staging of the disease. Bronchoscopy, gastroscopy, pharyngoscopy and microlaryngoscopy were part of staging for all patients in order to identify secondary malignancies, as well as regional and distant metastases. After staging, patients received therapy according to the primary tumor stage. Multimodal therapy included tumor resection, neck dissection, adjuvant chemoradiotherapy, or definitive chemoradiotherapy when surgical therapy was refused or deemed not suitable. All patients underwent regular follow-ups at the clinic, including head and neck examination, ultrasonography of the neck, microlaryngoscopy, esophagogastroscopy and bronchoscopy, with a mean interval of 4.8 months (standard deviation, 1.8 months). Patient data were retrieved from the internal tumor database of our institution, patient files, medical reports, surgical reports, and interdisciplinary tumor board protocols. Information was also obtained on patient sex, age, TNM stage, date of diagnosis, localization of the primary tumor, secondary carcinomas and their localization (synchronous and metachronous), and tobacco and alcohol consumption. The initial hypothesis was that no more than 20% of all cases would be positive for a secondary carcinoma. Another goal of the present study was to investigate possible clinical indicators for the occurrence of secondary carcinomas.

Statistical analysis

Statistical analysis was performed with JMP 11 software (SAS Institute, Inc.). A t-test was performed for independent variables. P≤0.05 was considered to indicate statistically significant differences.

Results

Patient characteristics

The majority of the 380 patients were male (n=269; 70.8%). The mean age was 63.5 years, with a standard deviation of 10.9 years (range, 24-94 years). The majority of the patients were diagnosed with laryngeal carcinoma (n=133; 35%), followed by 128 (33.7%) patients with oropharyngeal cancer. A total of 67 patients (17.6%) had carcinoma of the oral cavity. The distribution of tumor localization is shown in Table I.

Table I Distribution of primary tumor localization.

Table I

Distribution of primary tumor localization.

Localization	Frequency	%
Larynx	133	35
Oropharynx	128	33.7
Oral cavity	67	17.6
Hypopharynx	52	13.7
Total	380	100.0

The 7th edition of TNM classification was applied for cancer staging (36). The most common T stage was T2 (32.6%), followed by T1 (29.5%). The most common nodal status was N0 (46.8%), followed by N2 (41.1%). The majority of the patients had no distant metastases (M0; 91.8%). The distribution of TNM stage of the primary tumors is shown in Table II.

Table II Distribution of TNM stage of primary tumors within the cohort.

Table II

Distribution of TNM stage of primary tumors within the cohort.

A, T stage
Stage	Frequency	%
Carcinoma in situ	7	1.8
T1	112	29.5
T2	124	32.6
T3	76	20
T4	61	16.1
Total	380	100.0
B, N stage
Stage	Frequency	%
N0	178	46.8
N1	32	8.4
N2	156	41.1
N3	10	2.6
Nx	4	1.1
Total	380	100.0
C, M stage
Stage	Frequency	%
M0	349	91.8
M1	12	3.2
Mx	19	5
Total	380	100.0

Localization of secondary carcinomas

Between 2011 and 2015, a total of 39 patients (10.3%) developed a secondary carcinoma. Secondary carcinomas exhibited a similar distribution pattern to that of the primary tumors. The most common localizations of the secondary malignancies were the oropharynx (41%), larynx (25.6%) and oral cavity (20.6%). The localizations of the secondary carcinomas diagnosed in the patients of the present study are summarized in Table III.

Table III Distribution of secondary carcinoma localization.

Table III

Distribution of secondary carcinoma localization.

Localization	Frequency	%
Oropharynx	16	41
Larynx	10	25.6
Oral cavity	8	20.6
Hypopharynx	5	12.8
Total	39	100.0

Synchronous and metachronous secondary cancers

Secondary malignancies were differentiated into synchronous and metachronous. The time of diagnosis of the secondary carcinoma was used to distinguish between the two patient cohorts. A total of 22 (5.8%) of the patients developed a synchronous tumor (diagnosed within the first 6 months after the primary tumor), whilst 17 patients (4.5%) developed a metachronous tumor (occurring between 6 months and 5 years after the primary tumor). Figure 1 displays the number of diagnosed secondary malignancies (synchronous and metachronous) within the collective between 2006 and 2015. Whilst there was an increase in diagnosed secondary malignancies from 2006 to 2012, the number reached a steady state between 2012 and 2013 and decreased between 2014 and 2015.

Figure 1 Number of diagnosed secondary malignancies annually.

Associations between clinicopathological characteristics and secondary cancer development

A Chi-square test and a Fisher's exact test were performed to analyze the associations between different primary tumor localizations and the development of secondary carcinomas. A P-value of 0.1 confirmed There was no significant association between the localization of the primary tumor and the development of a secondary carcinoma (P=0.1), although the majority of secondary carcinomas affected patients with primary oropharyngeal (n=16; 4.2% of all patients; 41% of all secondary carcinomas) and laryngeal carcinomas (n=13; 3.4% of all patients; 25.6% of all secondary carcinomas).

Whether patient sex was a relevant factor for the occurrence of secondary carcinomas was investigated. No significant correlation between sex and secondary malignancies was observed (P=0.51), although more men (31; 8.2% of all patients) rather than women (8; 2.1% of all patients) developed a secondary carcinoma. The distribution of secondary malignancies amongst men and women is summarized in Table IV.

Table IV Distribution of secondary carcinoma amongst male and female patients.

Table IV

Distribution of secondary carcinoma amongst male and female patients.

Secondary carcinoma	Male	Female	Total
No, n (%)	269 (70.8)	72 (18.9)	341 (89.7)
Yes, n (%)	31 (8.2)	8 (2.1)	39 (10.3)
Total	300 (78.9)	80 (21.1)	380 (100.0)

The youngest patient in our study population with a secondary carcinoma was 50 years old, while the oldest was 87 years old. A t-test was performed in order to evaluate patient age as a possible risk factor for the development of secondary carcinomas. A P-value of 0.27 (95% CI: -1.07, 3.7) could not prove this hypothesis, which suggests that the development of secondary malignancies is not significantly associated with patient age.

A total of 42% of all secondary carcinomas had a T1 primary tumor and 26% had a T2 primary tumor (P=0.39). A total of 48% of the patients had a lymph node status of N0, and another 48% had N2a, b or c disease (P=0.39). Of all secondary tumors, 98% had no distant metastasis at the time of diagnosis (P=0.22). Chi-squared tests proved that the TNM status of the primary tumors did not represent a risk factor for the development of a secondary carcinoma.

A total of 138 patients reported regular consumption of carcinogenic substances, such as tobacco products, or alcoholic beverages; a total of 17.4% (70 patients) reported tobacco use alone, 5.1% (20 patients) reported alcohol use alone, and 12.2% (48 patients) reported use of both alcohol and tobacco. Two of the patients (0.5%) reported other forms of drug consumption. No information regarding the consumption of carcinogenic substances was available for the remaining 242 patients. A Chi-square test was performed to evaluate the association between consumption of carcinogenic substances and the development of secondary malignancies. Due to incomplete data regarding the substance abuse profile of 242 patients, no reliable general conclusion can be drawn from this statistical analysis. Although no significant association could be confirmed (P=0.89), a larger cohort with greater data availability should be investigated. A summary of the aforementioned statistical results is presented in Table V.

Table V Association between the development of secondary carcinoma and different cofactors.

Table V

Association between the development of secondary carcinoma and different cofactors.

Cofactors	P-value
Primary tumor location	0.1
Sex	0.51
Age	0.27
T stage of primary tumor	0.39
N stage of primary tumor	0.39
M stage of primary tumor	0.22
Continued exposure to carcinogenic substances	0.89

Discussion

The present study was designed to investigate indicators for secondary carcinomas in patients with head and neck cancer. The data of 380 patients with head and neck cancer were assessed to retrospectively analyze the incidence and frequency of synchronous and metachronous secondary carcinomas.

Patients with head and neck cancer are prone to develop multiple primary tumors (5,6). A possible explanation for this phenomenon is the concept of field cancerization, coined by Jaiswal et al in 1953(21). The reported incidence of secondary malignancies ranges between 1.8 and 34.6% (8-19). We were able to confirm our hypothesis that secondary head and neck malignancies occur in <20% of the patients in the present study. Of the 380 patients, 39 developed a secondary malignancy (10.3%). Increasing numbers of secondary carcinomas were detected between 2006 (n=1) and 2012 (n=13). After 2012, the number of secondary carcinomas stabilized and then declined again in 2014. During this period, our hospital established a certified head and neck cancer center, which may explain the growing number of diagnosed secondary carcinomas due to structured follow-ups, followed by a decline as a result of multi-disciplinary therapy. Furthermore, the establishment of withdrawal programs may have had a positive impact on the incidence of secondary malignancies (37-39). The pattern of anatomical distribution for secondary carcinomas was similar to that for primary tumors. The most common localization was the oropharynx (32%) followed by the larynx (20%) and oral cavity (16%). The most common localization of secondary malignancies varies in the current literature, and includes the lung, larynx and oropharynx (2,13,11). However, different follow-up standards were used. In the present study, all patients underwent a chest X-ray and a full endoscopic evaluation. Wax et al compared a computed tomography (CT) scan of the chest with simple X-ray and bronchoscopy. The CT scan exhibited the highest sensitivity at 87% (40). Another study reported that early-stage secondary malignancies can barely be detected on CT scans, so this examination should only be used in advanced-stage secondary carcinomas (stage III and IV) (41). Bronchoscopy is an invasive procedure, yet it allows biopsies and is associated with no exposure to radiation (14). In our cohort, only a small number (n=2; 5%) of secondary carcinomas were successfully diagnosed by bronchoscopy. According to current evidence, metachronous carcinomas are found more often compared with synchronous carcinomas (20). Depending on the study, the incidence varies between 4.7 and 24% for metachronous tumors, whereas synchronous secondary carcinomas are reported in 0.3-14% of the patients (9,11,13,15,20). In the present study, 6.2% of the cohort were found to have a synchronous tumor and 4.1% were found to have a metachronous tumor. The long interval of 6 months and 5 years leaves room for detection bias. Not all patients comply to the full scale of the follow-up program, whereas some patients relocate and cannot be evaluated to the full extent. Although the cohort of the present study cannot match a multicentered international cohort, the aim of this study was to evaluate the occurrence of secondary carcinomas within the confined environment of a single treatment center and, thus, help predict numbers and occurrences within smaller cohorts. The present study further aimed to evaluate the quality of data collected during the establishment of a certified tumor center.

As regards demographic patient characteristics, the mean age of the cohort was 64 years, and 78.9% of the patients were men, which is consistent with other studies (8,42). Malignancies in the head and neck area were more common among patients aged 60-70 years and significantly more frequent in men (n=31; 79.5% of all secondary carcinomas) rather than women (n=8; 20.5% of all secondary carcinomas). A t-test revealed that the association between age and secondary malignancy development in our study was not statistically significant (P=0.27, 95% CI: -1.08, 3.74). The sex predilection may be explained by the higher consumption of noxious substances by men. Similar studies report men to be more prone to the development of secondary malignancies (11,20). The majority of patients in our cohort were diagnosed with a primary T2 (31.98%) or T1 tumor (28.93%), which confirms the findings reported by Ruback et al (30). A similar distribution of nodal and metastatic status was found in a cohort formed by the University of Tokyo (9). Patients are more likely to recognize lesions and abnormalities in the upper airway at an early stage. However, non-specific symptoms, such as hoarseness, which may be initially attributed to a common cold, may delay primary diagnosis. Regardless, 60% of all laryngeal malignancies are diagnosed at an earlier stage compared with subglottic or supraglottic malignancies (28). Tumors of the oral cavity may also be detected at an early stage. Of the 112 cases, 88.4% had a tumor size of T1 or T2(12). A retrospective study of 52 patients with secondary carcinomas in the head and neck area conducted by Saito et al also found the index tumor size to be T1 or T2 in 62% of the cases (9). However, TNM classification was not found to be a significant indicator for the occurrence of secondary carcinomas in the present study.

A secondary tumor most often developed among patients with primary oropharyngeal carcinoma (16 cases; 4.2% of all patients; 41% of all secondary carcinomas), although oropharyngeal tumors were not the most common primary tumors. Patients with laryngeal cancer ranked second regarding the occurrence of secondary carcinomas (13 cases; 3.4% of all patients; 33% of secondary carcinomas). The lowest incidence of secondary carcinomas was observed for primary oral cavity and hypopharyngeal cancers (Table III). Dequanter et al reported the larynx, hypopharynx and oropharynx (in descending order) as the most common sites of secondary carcinoma development (43); Patrucco et al confirmed these results (11). In a large study with 58,363 carcinomas of the head and neck area, the majority of secondary carcinomas were diagnosed in patients with a primary carcinoma of the oral cavity (967 cases), larynx (353 cases), oropharynx (281 cases) and hypopharynx (76 cases) (8). Thus, there is an inconsistency in the current literature regarding the association between the localization of the primary tumor and secondary carcinomas (44,45). We were unable to identify a significant association between the localization of the primary carcinoma and the development of secondary carcinomas in the present study (P=0.09).

The combined abuse of tobacco and alcohol increases the risk of SCC (30). A retrospective study by Ruback et al reported consumption of tobacco and alcohol by 1,351 patients with malignancies in the head and neck area. A total of 747 patients (75.15%) reported regularly consuming tobacco, 579 (58.25%) alcohol, and 547 (54%) both (30). Another publication by Leon et al categorized 88.9% as smokers and 87.7% as regular alcohol consumers (n=4,298) (44). In the present study, 142 of the 380 patients reported regular alcohol and/or tobacco consumption. Tobacco use was more frequent (18.27% of patients), whereas 48 patients (12.18%) reported regular use of both alcohol and tobacco. In comparison to the available evidence, the number of patients consuming noxious substances in our data is low and was only obtained from 142 patients due to insufficient patient history, or patients not admitting to substance abuse. However, our results support the fact that nicotine and alcohol have carcinogenic properties. A total of 18.3% of patients without and 18% with secondary carcinomas were regular smokers. Regular smokers and drinkers developed secondary carcinomas more often compared with those who did not consume both substances regularly. Our data were not sufficient to establish a statistically significant effect of smoking and drinking on the development of secondary carcinomas. A study from the Head and Neck Center in St. Pau investigated 3,631 patients and confirmed that continued smoking and/or alcohol consumption after primary therapy significantly increased the risk for secondary tumors (P<0.001) (45). It is a matter for discussion whether invasive or mutilating procedures, such as laryngectomy, may trigger changes in drinking or smoking habits, since the majority of our patients with secondary carcinomas had T1 or T2 primary tumors. Eichler et al investigated 359 patients after laryngectomy; 68.5% of the patients quit smoking completely, 10.6% continued smoking, 6.1% had not been smokers before, whereas no data could be obtained for 14.8% of the cases. Regarding alcohol consumption, 28.1% of the patients continued alcohol consumption, whilst 70.5% tried to reduce consumption (42). Alcohol appears to have a higher addictive factor than nicotine; however, a more aggressive surgical approach does not appear to significantly affect the patient's behavior for either noxious substance.

Infection with high-risk types of human papillomavirus (HPV) is a well-known risk factor for the development of multiple carcinomas (8-10,30). In the present cohort, no data could be acquired regarding the HPV status or cumulative dose of radiation for the primary tumors. However, several studies suggest that infection with high-risk HPV types 16 and 18 increases the risk of development of secondary carcinomas. A study by Li et al from 2018 stated that HPV-infected patients are generally immunosuppressed and, thus, may be predisposed to the development of secondary cancers (46). However, a current meta-analysis by Götz et al, investigating the diagnosis of HPV infection, reported a widely varying number of infections with HPV amongst cohorts, and concluded that previously published studies regarding this subject should be read critically and may not represent a basis for therapeutic decisions (47). Further immunosuppression in patients with HNSCC occurs during chemoradiation (48). Al-Taei et al observed an overall decreased T-cell response and accumulation of immunosuppressive effects in oropharyngeal cancer patients following chemoradiation (49). As regards chemoradiation alone, a study by Berrington de Gonzalez et al demonstrated a significantly increased risk of developing secondary carcinomas in patients who received a radiation dose of >5 Gy (50). Since HNSCC patients regularly receive radiation with a cumulative dose of 40-60 Gy, these findings should be investigated further. Overall, the combination of virus- and treatment-induced immunosuppression may increase the risk of multiple carcinomas (51).

The primary limitation of this study was its retrospective design. The data on the consumption and abuse of noxious substances was particularly inconsistent and, thus, must be interpreted with caution. Another issue was recall bias: Treatment histories, medications and use of noxious substances, could not be fully recalled by the patients and physicians. When comparing the cohort of the present study to those of similar studies, it was found to be a representative sample. However, increasing the cohort size may help achieve statistically significant results in the case of smaller differences between groups. Further studies should collect data over a longer period of time, producing more consistent and detailed information. In conclusion, the study cohort is comparable to those of other studies, and appears to be a representative sample of head and neck cancer patients. The incidence of secondary carcinoma in this study was 10.3%. The localization pattern of secondary carcinomas was similar to that of primary tumor sites. The majority of secondary carcinomas were found in the oropharynx (32%), followed by the larynx (20%) and the oral cavity (16%). These results are in line with the concept of field cancerization, suggesting that secondary carcinomas may develop near the primary tumor site. However, the results of the present study were not statistically significant. Synchronous tumors occurred slightly more often compared with metachronous tumors (6.2 vs. 4.1%, respectively). Age and sex did not significantly affect the occurrence or development of secondary carcinomas. Furthermore, no significant association was identified between the TNM status of the primary cancer and that of the secondary carcinoma. Regarding the consumption of noxious substances, a clear trend was observed in that persistent consumption of alcohol and tobacco after the primary diagnosis promoted the development of secondary carcinomas. Yet, due to the inconsistencies in the data collected from 252 patients in the cohort, the statistical significance of this association was not proven (P=0.89). Overall, avoiding the continued exposure to noxious substances, as well as early diagnosis and consistent treatment in a certified head and neck cancer center, appear to be key factors for decreasing the incidence of secondary carcinomas. Furthermore, the role of infection with high-risk HPV types, as well as the impact of chemoradiation, should be investigated further, as they appear to play a key role in the development of secondary malignancies.

Acknowledgements

The authors would like to thank Mrs. Sylvia Büttner, of the Institute for Medical Statistics, Biomathematics and Information Processing in Mannheim, for creating the statistical breakdown.

Funding

No funding was received.

Availability of data and materials

Data were obtained from the oncological database for head and neck cancer at the ENT Tumor Center Mannheim.

Authors' contributions

VN analyzed and interpreted the data retrieved from the oncological database and was a major contributor to literature research. BKu interpreted the data and wrote the manuscript. BKr, NR and CA were major contributors in correcting and proofreading the final manuscript. CA and NR helped with acquiring the data. Further, NR and CA have given final approval of the version to be published. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The study protocol was approved by the Ethics Committee II of the Medical Faculty of Mannheim at the University of Heidelberg, Mannheim, Germany (file no. 2016-827R-MA).

Patient consent for publication

Not applicable.

Competing interests

All the authors declare that they have no competing interests.

References