Does Time to Initiation of Adjuvant Radiotherapy Affect Reconstruction Outcomes after Endoscopic Resection of Skull Base Malignancies?

Abstract

Introduction and Objective It is unclear if the length of the time interval to initiation of adjuvant radiation therapy (RT) after endoscopic endonasal surgery affects reconstruction outcomes. In this study we present our experience with adjuvant RT after endoscopic endonasal procedures, to determine if the time to RT after surgery impacts post-RT reconstruction complication rates.

Methods A retrospective cohort study of 164 patients who underwent endoscopic endonasal surgery between 1998 and 2021 was conducted. Using Cox proportional hazard ratios (HRs), we evaluated several variables and the complications that occurred during the 1-year period after starting RT.

Results Seventy-eight (47.5%) and eighty-six patients (52.5%) received RT before and after the sixth postoperative week, respectively. The overall post-RT complication rates were 28%, most of these were severe infections (n = 20, 12.2%) and delayed CSF leak (n = 4, 2.5%). There was no significant difference in the post-RT complications between the patients who received postoperative RT before or after the sixth operative week (HR: 1.13; 95% confidence interval: 0.63–2.02; p = 0.675). Univariate analysis demonstrated negative impact associated with smoking history (p = 0.015), the use of neoadjuvant chemotherapy (p = 0.0001), and the use of photon therapy (p = 0.012); and we found a positive impact with the use of multilayer reconstruction techniques (overall, p = 0.041; with fat, p = 0.038; and/or fascia graft, p = 0.035). After a multivariate analysis only, smoking history was an independent risk factor for post-RT complications (p = 0.012).

Conclusion Delaying RT for more than 6 weeks after endoscopic endonasal surgery does not provide a significant benefit for reconstruction outcomes. However, special attention may be warranted in patients with smoking history who have received neoadjuvant chemotherapy, or in patients who will receive photon-based RT after surgery as these groups were found to have increased complication rates post-RT.

Keywords

endoscopic skull base reconstructions - vascularized flap - radiation therapy - proton therapy - photon therapy

Introduction

Malignancies of the skull base pose both inherent surgical challenges and overall challenges with management with regard to multimodal therapy. Beyond maximizing oncologic goals, any surgical strategy must ensure timely recovery and initiation of adjuvant therapies. In malignant sarcomas or sinonasal tumors, radiation therapy (RT) is critical in the management of the tumor margins and focus must be placed on the radiation target volumes.[1] In practice, there is a general acceptance to avoid unnecessary delays in RT.[2] Delaying RT has shown to be significantly associated with increased risk of local recurrence in multiple cancers.[3] The proliferation of clonogenic cells in the cancer cell environment[4] permits the spread of the cancer within and potentially beyond the radiation treatment volume and thus may ultimately impact clinical outcomes. Delaying RT in sinonasal cancers in particular for more than 50 days was associated with worse overall survival.[5] Therefore, it was recommended that care teams should focus on shortening the time to RT as much as possible to improve the outcomes without increasing the risks of early radiation.

One of the main concerns and risks of adjuvant RT is interruption of the normal healing of the surgical wound.[6] RT impacts not only dividing cancer cells, but also other rapidly dividing cells found in the skin, mucosa, or the epithelial lining of the sinonasal cavity.[7] [8] It is well accepted that maturation and remodeling of the scar tissue takes at least 2 weeks after injury and may last up to 1 year,[8] so a surgical wound should not be radiated within the first 2 weeks. Most endothelial cells, macrophages, and fibroblasts undergo apoptosis or exit the wound. Furthermore, and particularly in the sinonasal area, radiation damages the respiratory epithelium and induces squamous cell metaplasia, impairing the normal mucociliary clearance.[9] [10] This combination of impaired wound healing and increased risk of infection should be balanced with clinical benefit to determine an optimal time to begin RT.

With the increased use of endoscopic approaches in the management of skull base malignancies, the procedure length, the postoperative recovery, and associated adverse events have decreased compared to open cranial approaches.[11] Nevertheless, it is unclear if time to initiation of adjuvant RT affects reconstruction outcomes after endoscopic endonasal surgery and whether a delay in starting RT should be considered. This question is important, because the delay in RT initiation may counter the benefits of endoscopic surgery.

We conducted a retrospective analysis of data from patients who received adjuvant RT after endoscopic endonasal procedures for skull base malignancies and evaluated whether the length of the time interval from surgery to initiation of RT impacted the post-RT reconstruction complication rates.

Methods

We performed a retrospective study using data from patients treated at our institution between January 1998 and July 2021. The clinical data from patients of any age with skull base tumors with intracranial extension treated with postoperative radiation after endoscopic endonasal surgery were collected from their medical records. The medical records of patients had to include complete information regarding patient medical history, magnetic resonance imaging (MRI) studies, surgical and closure techniques, details of hospitalization, and postoperative radiation plan. Our institutional review board approved this study for the review of the clinical records under Protocol 2023-0073; this study was deemed to not require patient consent. After review, 164 patients met inclusion criteria and patient-, surgery-, and treatment-related variables were gathered from clinical notes and operative reports.

Demographic variables included age, sex, and systemic wound healing risk factors[12] (SWHFs). The SWHFs included the following: smoking history, which was defined as regular use of cigars or cigarettes in the past 2 years[13]; obesity, defined as body mass index of 30 or higher[14]; presence of metabolic disease, including diabetes, hereditary healing disorder, jaundice and uremia; presence of a cardiovascular risk factor, including high blood pressure, ischemic heart disease, peripheral ischemic disease, or congestive heart failure; presence of hypoproteinemia, confirmed with preoperative low serum albumin; use of certain medication, such as the chronic use of steroids and current chemotherapy; presence of immune system disease, either primary or secondary immunosuppression; and alcoholism, defined as the daily consumption of an alcoholic beverage.

For statistical purposes, tumor histology was divided into bony sarcomas (chordomas, chondrosarcomas, and osteosarcomas), epithelial malignancies (esthesioneuroblastoma, squamous cell carcinoma, neuroendocrine carcinoma, sinonasal adenocarcinoma, malignant melanoma, and sinonasal undifferentiated carcinoma), soft tissue sarcomas (rhabdomyosarcoma, Ewing sarcoma), and other tumors (pituitary tumors, meningiomas, and metastasis).

Imaging was assessed for tumor location, the anatomic zone approached, and the extent of resection. Tumor location was defined according to the cranial fossa containing the whole tumor (or more than the half of the tumor mass when the tumor involves two cranial fossae). Tumors in sellar/suprasellar location were considered a separate category. The types of anatomic zone approaches were divided according to a Zone-Based Classification System in median, paramedian, or both zones.[15] The extent of resection was defined according to the postoperative MRI: no macroscopic residual tumor volume was considered as a gross total resection (GTR); postoperative residual tumor volume or positive tumor margins were considered as subtotal resection; significant tumor remnants according to the postoperative MRI and the operative note were considered partial resections or biopsies. The intraoperative cerebrospinal fluid (CSF) leak was categorized into four levels[16] and dichotomized for statistical purposes. Variables related to reconstruction included the use of a multilayer closure, the use of a vascularized or free flap, and the type of vascularized flap, the use of fat or fascia grafts and the use of synthetic dural tissue.

Time since surgery (in days) to the first radiation dose for each patient was collected and patients were divided into two groups for analysis: those who received RT more than 6 weeks after surgery and those who received RT less than 6 weeks after surgery (the median was 43 days). Information regarding postoperative flap ischemia, mucocele, or extensive crusting was collected from the postoperative otolaryngology-head and neck surgery outpatient notes and before starting RT (median time was 22 days). The types of RTs included photon-based RT (external beam RT and/or stereotactic radiosurgery) and proton-based RT. Total dose and number of fractions used in RT were collected for analysis.

The primary outcome assessed was complication-free survival after RT (RT-CFS). It was defined as the time from the first radiation fraction dose received to the presence of a sinonasal complication in the year following RT. A complication was defined as any kind of reconstruction or sinonasal issue requiring either inpatient or outpatient management.

Statistical Analysis

All statistical analyses were performed with SPSS 26 (IBM Corp, New York, United States) and Graph Pad 9.0 (GraphPad Software, California, United States). We used descriptive statistics, and presented findings as a mean ± standard deviation, for patient demographic data, tumor characteristics, intraoperative findings, and treatment-related variables. Statistical analysis was performed to determine the association of these variables with the primary endpoint. Chi-square test with Yates correction and Fisher exact test were used for nominal variables, and Student's t-test with Levene's test for normality was used to compare the means of age, total radiation dose, and number of fractions.

The RT-CFS curves were generated using the Kaplan–Meier method. The log-rank test was used to evaluate the relationship between RT-CFS and the different clinical, surgical, and treatment parameters. The univariate Cox-proportional hazards regression model was used to identify those variables associated with an increased risk of complication and each hazard ratio (HR) was reported along with a 95% confidence interval. Factors with a p-value less than 0.05 in the univariate Cox regression model were further included in multivariate Cox regression model by using a backward elimination procedure.

Results

Patient Demographics and Tumor Features

One hundred and sixty-four patients were treated with fully endoscopic procedures or combined transcranial and endoscopic approaches ([Table 1]). The median time to RT was 43 days (interquartile range [IQR]: 37–58, range: 13–240 days) with a skewness of 3.15; in our cohort we had 78 patients who received RT earlier than 6 weeks after surgery and 86 patients who received RT later than 6 weeks.

Table 1 Demographic features

Variables	Total (N = 164)	Less than 6 weeks to first RT dose (N = 78)	More than 6 weeks to first RT dose (N = 86)	p-Value (X2, Fisher's or t-test)
Preoperative
Age at surgery	49 ± 17	46 ± 16	52 ± 17	0.01
Sex
Female	76 (46.4%)	31 (39.7%)	45 (52.3%)	0.107
Male	88 (53.6%)	47 (60.3%)	41 (47.6%)
Histology
Bony sarcomas	68 (41.5%)	19 (24.4%)	49 (57%)	0.0001
Epithelial malignancies	86 (52.4%)	52 (66.7%)	34 (39.5%)	0.001
Soft tissue sarcomas	2 (1.2%)	1 (1.2%)	1 (1.2%)	–
Others	8 (4.9%)	6 (7.7%)	2 (2.3%)	–
Cranial localization
Anterior Fossa	84 (51.2%)	51 (65.4%)	33 (38.4%)
Sellar/suprasellar	10 (6.1%)	5 (6.4%)	5 (5.8%)	0.004
Middle fossa	12 (7.3%)	3 (3.8%)	9 10.5%)
Posterior fossa	58 (35.3%)	19 (24.3%)	39 (45.4%)
Systemic wound healing risk factor
Cardiovascular risk factors	76 (46.3%)	33 (42.3%)	43 (0.5%)	0.324
Metabolic disorder	60 (36.6%)	28 (35.9%)	32 (37.2%)	0.298
Hypoproteinemia	1 (0.6%)	0 (0%)	1 (1.2%)	1.00
BMI > 30	37 (22.5%)	12 (15.3%)	25 (29%)	0.036
Smoking	43 (26.2%)	19 (24.3%)	24 (27.9%)	0.606
Current medication	8 (4.9%)	4 (5.1%)	4 (4.6%)	1.00
Alcoholism	9 (5.5%)	3 (3.8%)	6 (7%)	0.50
Immunocompromised	9 (5.5%)	6 (7.6%)	3 (3.5%)	0.311
History of previous radiation	14 (8.5%)	4 (5%)	10 (11.6%)	0.168
History of previous surgery	39 (23.8%)	17 (21.8%)	22 (25.6%)	0.569
Use of Neoadjuvant chemotherapy	17 (10.3%)	9 (11.5%)	8 (9.3%)	0.639
Surgery and reconstruction techniques
Type of expanded approach
Medial sagittal plane	127 (77.4%)	66 (84.6%)	61 (70.9%)	0.106
Paramedian sagittal plane	17 (10.3%)	6 (7.7%)	11 (12.8%)
Combined	20 (12.2%)	6 (7.7%)	14 (16.3%)
Size of intraoperative CSF leak
Low flow (Grade 1 or 2)[a]	60 (36.5%)	30 (38.4%)	30 (35%)	0.635
High flow (Grade 3 or 4)[b]	104 (63.5%)	48 (61.5%)	56 (64.5%)
Extent of resection
GTR	130 (79.3%)	63 (80.7%)	67 (78%)	0.652
STR	32 (19.5%)	13 (16.6%)	19 (22%)
Partial or biopsy	2 (1.2%)	2 (2.5%)	0 (0%)
Use of vascularized flap	123 (75%)	59 (75.6%)	64 (74.4%)	0.857
Use of fascia graft	81 (49.4%)	31 (39.7%)	50 (58%)	0.019
Use of fat graft	27 (16.5%)	7 (9%)	20 (23.2%)	0.014
Postoperative management
Postoperative CSF leak	18 (10.9%)	5 (6.4%)	13 (15%)	0.085
Postoperative meningitis	8 (4.9%)	2 (2.5%)	6 (7%)	0.282
Type of RT used
EBRT	79 (48.1%)	48 (61.5%)	31 36%)	0.006
PBRT	74 (45%)	28 (35.9%)	46 (53.5%)
SRS	10 (6%)	2 (2.5%)	8 (9.3%)
EBRT + SRS boost	1 (0.6%)	0 (0%)	1 (1.2%)
Number of fractions	28 (2-39)	28 (2-39)	29 (1-39)	0.89
Total dose	58.5 ± 14.4 Gy	57.6 ± 13 Gy	59.2 ± 15.6 Gy	0.475
Use of adjuvant chemotherapy	22 (13.4%)	12 (15.4%)	10 (11.6%)	0.481

Abbreviations: BMI, body mass index; CSF, cerebrospinal fluid; EBRT, external beam radiotherapy; PBRT, proton beam radiotherapy; RT, radiation therapy; SRS, stereotactic radiotherapy.

Note: Bold/italic letters indicate variables with statistical significance.

^a Grade 1: no visible or minor with Valsalva; Grade 2: minor leak evident with Valsalva maneuvers.

^b Grade 3: clear dural defect with moderate/severe persistent leak; Grade 4: high-flow leak predicted to occur as part of the expanded approach, or tumor larger than 3 cm.

The presence of a cardiovascular disease (46.3%), a metabolic disease (36.6%), smoking habit (26.2%), and obesity (22.5%) were the most frequent wound healing risk factors reported. Overall, 37 patients (22.7%) had the presence of three or more SWHFs when receiving RT.

Epithelial malignancies were the most frequent tumors reported (52.4%), with primary location in the anterior fossa (94.2%, n = 81). Bony sarcomas represented 41.5% of all cases and were located mostly in the posterior fossa (80.9%, n = 55), followed by the middle fossa (13.2%, n = 9). A larger proportion of patients with bony sarcomas received RT after the sixth postoperative week (p = 0.0001) and, conversely, patients with epithelial malignancies received RT before the sixth postoperative week (p = 0.001).

The use of neoadjuvant induction chemotherapy was reported only in primary sinonasal tumors (18.5%, n = 17). Among them, esthesioneuroblastoma and malignant melanoma represent five cases each. History of previous RT was reported in 14 patients (8.5%), including eight chordomas, four epithelial malignancies, one radiation-induced osteosarcoma, and one rhabdomyosarcoma.

Surgical Management

The tumors were located in the anterior fossa (51.2%, n = 84), posterior fossa (35.4%, n = 58), middle fossa (7.3%, n = 7.3%), and in the sellar/suprasellar region (6.1%, n = 10). For anterior fossa tumors an exclusive approach to the medial sagittal plane (transcribriform/transplanum) was used in 89% of the cases (n = 75) and a paramedian extension (transorbital) was required in 11% (n = 9) of the cases. Conversely, the paramedian extension (via transpterygoid approach) was more frequently required in posterior fossa tumors (43%, n = 25) and the exclusive medial sagittal plane (transclival/transodontoid) approaches were used in 24% (n = 14).

In terms of resection, GTR was achieved in 79.3% (n = 130) of the cohort, and after resection, a clear dural defect/high-flow leak was present in 63% of the cases (n = 104). According to the size of the intraoperative CSF leak, there was no leak in 36 (22%) cases, 24 cases had a leak only with Valsalva maneuvers (14.6%), 31 cases had minor dural tears (18.9%), and 73 cases had a significant dural defect or a resection of a subdural component larger than 3 cm (44.5%).

For reconstruction, a multilayer closure technique was used in 66% (n = 108) of the cases. Multilayer closure was used mainly in posterior fossa tumors (79%, n = 47 vs. 56%, n = 46; p = 0.007). Fat and fascia grafts were used primarily for reconstruction of the posterior fossa, in 40% (n = 23) and 63.7% (n = 37) of cases, respectively. In the anterior fossa, the use of grafts was significantly lower (p = 0.002), fat was rarely used (3.7%, n = 1), and fascia was used in 36.9% (n = 31) of cases.

The final layer was vascularized in 75% of the cases (n = 123). Among them, 89% (n = 93) had grade 3 to 4 CSF leaks located in the anterior or posterior fossa in similar frequencies (43 and 44%, respectively). Nasoseptal and pericranial flaps were the only vascularized flaps used in 56% (n = 70) and 44% (n = 53) of the cases, respectively. Nasoseptal flap was used in 78% of the posterior fossa but in 33% of the anterior fossa reconstruction, where pericranial flap was most frequently used (66%, n = 36).

Postoperative Complications

Among 30-day postoperative complications, CSF leak occurred in 10.9% (n = 18) and postoperative meningitis in 4.9% (n = 8). Twelve CSF leaks were resolved with a lumbar drain placement and six required an endoscopic surgical repair. Neither obesity nor smoking history were associated with increased risk of 30-day postoperative CSF leak (p = 0.24 and p = 0.46, respectively). Furthermore, the size of the dural defect was not statistically associated with the incidence of postoperative CSF leak (p = 0.073) but was associated with the incidence of postoperative meningitis (p = 0.028).

Postoperative flexible rhinoscopy was performed in 117 patients (71%) and after a median of 22 postoperative days (IQR: 15–31.25 days). Extensive crusting was the most frequently reported finding (36%), followed by flap ischemia (3%) and mucocele (3%). There was a significant association between extensive postoperative crusting and the presence of three or more systemic wound healing factors (p = 0.009).

Adjuvant Radiation Therapy

Proton-based therapy was used in 74 patients (45%). Of them, 28 (37.8%) received the therapy before the sixth postoperative week and 46 (62.2%) received it after sixth postoperative week. Photon-based therapy was used in 90 patients (55%), of which 50 (55.5%) and 40 (44.5%) received it before and after the sixth postoperative week, respectively. In the photon-based cohort, 69 patients had epithelial malignancies (93.2%) and, in proton-based therapy, 57 patients had bony sarcomas (84%).

The overall post-RT complication rates were 28% (n = 46) with more than half requiring hospitalization (59%, n = 27) ([Table 2]). The median time for an RT-related complication was 101 days (IQR: 45.75–144.75) with a 6-month RT-CFS of 15%. The mean total dose applied in patients with RT-related complications was 57.1 ± 10 Gy and was lower than the mean dose applied in patients without complications (59 ± 15.7 Gy).

Table 2 Postradiation complications

RT- related complications	Total (N = 164)	Less than 6 weeks to first RT dose (N = 78)	More than 6 weeks to first RT dose (N = 86)
Not requiring hospitalization	19 (11.5%)	7 (9%)	12 (14%)
Post-RT sinus infection (outpatient)	13 (7.9%)	5	8
Mechanical nasal obstruction	3 (1.8%)	1	2
Mechanical lacrimal duct obstruction	2 (1.2%)	1	1
Severe neuralgia	1 (0.6%)	0	1
Requiring hospitalization	27 (16.5%)	16 (20.5%)	11 (12.8%)
Post-RT sinus infection (inpatient)	10 (6%)	7	3
Delayed CSF leak	4 (2.4%)	0	4
Facial cellulitis	4 (2.4%)	2	2
Skull base osteomyelitis	3 (1.8%)	2	1
Orbital cellulitis	3 (1.8%)	3	0
Spontaneous epistaxis	3 (1.8%)	2	1
Factors related to delayed RT (>6 weeks after surgery)			Total (N = 86)
Not specified			43 (50%)
Long POP hospitalization			23 (26.7%)
Insurance/financial issues			11 (12.8%)
Outpatient clinical worsening/complication			7 (8.1%)
Successive surgical procedure			1 (1.2%)
New disease stage requiring new RT planning			1 (1.2%)

Abbreviations: CSF, cerebrospinal fluid; POP, postoperative; RT, radiation therapy.

The main reasons for hospitalization were severe infectious diseases (n = 20, 12.2%) and delayed CSF leak (n = 4, 2.5%). The severe infectious diseases included postradiation rhinosinusitis (PRS) (n = 10), osteomyelitis (n = 3), and cellulitis (n = 7). PRS was confirmed clinically (with flexible rhinoscopy) in those patients who were in the inpatient and outpatient settings, and with bacterial cultures in those patients requiring hospital admission. All of them occurred more frequently in patients treated with photon therapy instead of proton therapy (n = 14, 70%), with primary sinonasal malignancies (n = 17, 85%) and treated with anterior cranial fossa approaches (29.8 vs. 6.9% in posterior fossa; X2 [2, n = 129] = 12.7, p = 0.0003). The delayed CSF leaks occurred in three patients with chordoma and one patient with esthesioneuroblastoma. Although the delayed CSF leak was more frequently reported after approaches to the posterior fossa, this difference was not statistically significant compared with anterior fossa (p = 0.30). All closures were done with a vascularized flap (two with nasoseptal flap and two with pericranial flap). Three closures were in a multilayer fashion, using fat and fascia lata grafts and one was a monolayer closure (esthesioneuroblastoma). Two patients had a postoperative CSF leak requiring a new endoscopic closure before receiving the RT, but this closure failed again after RT, requiring a new surgical closure.

Among the possible causes for receiving RT after 6 weeks, 50% (n = 43) could not be determined, 26.7% (n = 23) had a hospital stay longer than a week, and 12.8% (n = 11) had insurance/financial problems.

Analysis of Factors Associated with RT-CFS

Upon conducting univariate log-rank and Cox proportional hazards analysis, we identified those variables that correlated with higher post-RT complication rate ([Table 3]). The analysis did not identify a statistically significant increased risk of complication in patients receiving radiation before the sixth postoperative week (HR: 1.13; 95% CI: 0.634–2.018; p = 0.675) ([Fig. 1A]). The regular use of tobacco in the past 2 years was the only SWHF that showed a significantly higher complication risk (HR: 0.489; 95% CI: 0.248–0.964; p = 0.0157) ([Fig. 1C]), which was confirmed upon multivariate analysis (p = 0.012), as seen in [Table 4]. Interestingly, the use of neoadjuvant chemotherapy was a negative prognostic factor in univariate (HR: 0.25; 95% CI: 0.086–0.773; p = 0.0001) ([Fig. 1D]), but lost statistical strength in a multivariate analysis (p = 0.052). Among demographics variables, the 6-month RT-CFS was 92 and 66% for bony sarcomas and epithelial malignancies, respectively (HR: 4.9; 95% CI: 2.69–8.91; p = 0.0001) ([Fig. 1B]), but this could not be replicated in a multivariate analysis (p = 0.382). Tumors located in the anterior fossa, where 96% (n = 81) were epithelial malignancies, showed significantly worse prognosis, with a 6-month CFS of 65% (p = 0.001). On the other hand, the 6-month CFS of posterior fossa tumors was 90%, where 95% (n = 55) were bony sarcomas ([Fig. 1E]).

Fig. 1 RT complication-free survival curves. (A) RT-CFS according to time to RT, p = 0.675. (B) RT-CFS according to histology, p = 0.0001. (C) RT-CFS according to smoking history, p = 0.0157. (D) RT-CFS according to use of neoadjuvant chemotherapy, p = 0.0001. (E) RT-CFS according to cranial localization of the tumor (anterior fossa vs. posterior fossa), p = 0.0002. RT, radiation therapy; RT-CFS, RT complication-free survival.

Table 3 Univariate analysis of risk factors in patients with or without post-RT complications

Risk factors	No post-RT complication (n = 118)	Post-RT complication (n = 46)	Univariate analysis (complication-free survival)
			6-month CFS	HR	95% CI	p-Value
Preoperative factors
Age (>50 vs <50)
• <50 yo	63	17	71%	1.69	0.948–3.01	0.0819
• >50 yo	55	29	81%
Sex (female vs. male)
• Female	57	19	79%	0.771	0.433–1.377	0.385
• Male	61	27	74%
Histology[a]
• Bony sarcomas	61	7	92%	4.90	2.69–8.91	0.0001
• Epithelial malignancies	50	36	65%
• Soft tissue sarcomas	1	1	50%
• Others	6	2	75%
Systemic wound healing factors
• 0–2 SWHF	93	34	78%	0.77	0.38–1.56	0.381
• ≥3 SWHF	25	12	70%
Smoking history (No vs. Yes)
• Yes	25	18	63%	0.489	0.248–0.964	0.0157
• No	93	28	81%
BMI (>30 vs. <30)
• >30	28	9	81%	1.24	0.625–2.46	0.351
• <30	90	37	74%
Previous radiation (No vs. Yes)
• Yes	9	5	71%	0.771	0.275–2.163	0.583
• No	109	41	76%
Previous surgery (No vs. Yes)
• Primary treatments	92	33	77%	0.795	0.403–1.566	0.48
• Recurrent treatments	26	13	74%
Neoadjuvant chemotherapy
• Yes	6	11	80%	0.258	0.086–0.773	0.0001
• No	112	35	41%
Surgery-related factors
Type of expanded approach[b]
• Medial sagittal plane	87	40	73%	7.19	3.06–16.8	0.02
• Paramedian sagittal plane	12	5	76%
• Combined	19	1	95%
Cranial localization[c]
• Anterior fossa	49	35	65%	4.12	2.247–7.58	0.0002
• Sellar/suprasellar	9	1	90%
• Middle fossa	9	1	75%
• Posterior fossa	51	7	90%
Size of intraoperative CSF leak
• No visible or minor with Valsalva	46	14	78%	0.738	0.406–1.343	0.341
• Clear defect or tumor >3 cm	72	32	75%
Extent of resection
• GTR	92	38	74%	1.326	0.658–2.668	0.467
• STR/partial	26	8	85%
Use of multilayered closure (No vs. Yes)
• Yes	83	25	81%	1.809	0.972–3.367	0.041
• No	35	21	66%
Use of vascularized flap for reconstruction (No vs. Yes)
• Yes	90	33	77%	1.196	0.612–2.336	0.58
• No	28	13	73%
Use of fascia graft for reconstruction (No vs. Yes)
• Yes	64	17	85%	1.881	1.055–3.355	0.035
• No	54	29	67%
Use of fat graft for reconstruction (No vs. Yes)
• Yes	24	3	93%	3.211	1.521–6.779	0.038
• No	94	43	73%
Postoperative factors
Postoperative CSF leak
• Yes	13	5	78%	0.957	0.372–2.461	0.925
• No	105	41	76%
Postoperative meningitis
• Yes	6	2	87%	1.161	0.308–4.36	0.836
• No	112	44	76%
Extensive crusting in postoperative rhinoscopy
• Yes	40	19	73%	0.773	0.422–1.416	0.388
• No	78	27	78%
Type of radiation applied
• Photon	57	33	70%	0.449	0.251–0.802	0.012
• Proton	61	13	84%
Use of adjuvant chemotherapy (No vs. Yes)
• Yes	15	7	73%	0.843	0.356–1.984	0.678
• No	103	39	77%
Time to RT after surgery (<6 wk vs. >6 wk)
• <6 wk	55	23	74%	1.131	0.634–2.018	0.675
• >6 wk	63	23	77%

Abbreviations: BMI, body mass index; CFS, complication-free survival; CI, confidence interval; CSF, cerebrospinal fluid.

Note: Bold/italic letters indicate variables with statistical significance.

^a Univariate log-rank and Cox proportional hazards were analyzed between epithelial malignancies and bony sarcomas.

^b Univariate log-rank and Cox proportional hazards were analyzed between approaches through the medial sagittal plane and combined.

^c Univariate log-rank and Cox proportional hazards were analyzed between anterior fossa location and posterior fossa location.

Table 4 Multivariate analysis of risk factors affecting RT-CFS

Clinical factors	Multivariate analysis
	HR	95% CI	p-Value
Histology	1.27	0.742–2.177	0.382
Smoking history (No vs. Yes)	2.216	1.18–4.13	0.012
Neoadjuvant chemotherapy (No vs. Yes)	2.136	0.993–4.593	0.052
Tumor location	0.819	0.563–1.191	0.297
Type of expanded approach	0.828	0.386–1.778	0.629
Multilayer closure (No vs. yes)	0.832	0.411–1.686	0.61
Use of fascia graft for reconstruction (No vs. Yes)	0.802	0.378–1.702	0.566
Use of fat graft for reconstruction (No vs. Yes)	0.737	0.197–2.756	0.65
Type of RT used (photon vs. proton)	0.826	0.369–1.848	0.641

Abbreviations: CI, confidence interval; HR, hazard ratio; RT-CFS, radiotherapy complication-free survival.

Note: Bold/italic letters indicate variables with statistical significance.

When looking at reconstruction variables, multilayer closure (HR: 1.8; 95% CI: 0.972–3.367; p = 0.041), the use of fat graft (HR: 3.21; 95% CI: 1.52–6.78; p = 0.038), and the use of fascia graft (HR: 1.88; 95% CI: 1.05–3.35; p = 0.035) were significantly associated with a lower complication rate, but none of them could be replicated in multivariate analysis.

Finally, there was a statistically significant higher rate of complications with the use of photon-based therapies compared to proton-based therapies (HR: 2.22; 95% CI: 1.247–3.97; p = 0.012), but again this could not be replicated in a multivariate analysis. When compared with neoadjuvant chemotherapy, adjuvant chemotherapy was not a risk factor for RT complications.

Discussion

This study was conducted to evaluate the RT-related complication rates in patients with endoscopic skull base surgeries who received adjuvant RT before or after the sixth postoperative week. Based on our findings, we concluded that there is no increased risk of RT-related complications in those patients receiving radiation within 6 weeks after surgery. Timing alone does not seem to be a risk factor, but special attention must be given to those patients with recent smoking history or diagnosis of a sinonasal malignancy. Also, patients who received treatment with neoadjuvant chemotherapy or had a single-layer skull base reconstruction seem to have an increased risk for RT-related complications.

Focusing on sequential treatments and the main outcome of this study, the time between the first and the last treatment is one of the most important predictors of survival.[17] Prolonged or delayed RT has proven to have detrimental impact on the prognosis of different cancers,[18] [19] [20] including sinonasal cancers.[21] Nevertheless, applying radiation too early has a detrimental effect in the normal healing process of the surgical wound[8] and in this sense, the time at which RT is initiated must be balanced. Although our cut-off time was determined according to the median time since surgery to the first RT dose, there is evidence to support giving RT around the sixth postoperative week, because this has a significant impact on the prognosis of sinonasal malignancies[11] and other head and neck cancers.[22] Furthermore, our cut-off was even lower compared with other reported timings in head and neck cancers[5] and is the time period recommended by the National Comprehensive Cancer Network.[23] With supporting evidence and our results, we conclude that delaying adjuvant radiation beyond the sixth postoperative week for healing purposes has no benefit in reducing the RT-complication rates and might have a negative effect in prognosis of skull base tumors.

Irradiation to the sinonasal cavity causes profound damage to the respiratory epithelium, predisposing radiated patients to sinonasal infections or chronic rhinosinusitis (CRS).[9] [10] Metaplasia and subepithelial edema are the main histologic findings.[24] The incidence of CRS in other case series has been observed to be up to 16.7%[25] and was similar in our cohort, with an incidence of 14% (n = 23) during the first year post-RT. Our cohort represents only patients receiving adjuvant RT but it was reported that the increased risk for CRS is similar in those patients receiving RT as primary or adjuvant therapy combined with surgery[26] with an incidence peak occurring at 6 to 9 months after the RT.[27] Our results are in agreement with these prior findings, with infections representing two-thirds of all the RT-related complications (72%, n = 33). The remainder of delayed complications included neuritis and mechanical obstructions related to scarring and late fibrosis.[28] [29] Prolonged crusting due to impaired mucociliary clearance was associated with scarring and fibrosis[30] and in our cohort, prolonged crusting was frequently reported; however, in contrast to observations from other studies,[31] we did not find a lower incidence of prolonged crusting when comparing proton-based with photon-based RT (p = 0.28).

Smoking history was the only risk factor that demonstrated an association with increased RT-related complication rate, and to our knowledge, this is the first study that has analyzed this topic in endoscopic skull base reconstructions. It is well known that cigarette smoking reduces the capacity for any wound repair, particularly for patients undergoing reconstructive surgery.[32] There are several documented mechanisms of delayed wound healing and most of them occur at a microvascular level, with decreased tissue oxygenation and impaired reparative cell function.[33] In head and neck reconstructive surgery, Crippen et al found that patients who smoked cigarettes in the year prior to undergoing a free flap reconstruction with microvascular anastomosis had almost twofold increase in the rate of wound dehiscence.[32] Conversely, two different studies specifically evaluated the risk factors associated with postoperative CSF leak after skull base reconstruction and found that smoking history was not associated with increased risk of postoperative CSF leak,[34] [35] which was replicated in our cohort, 30-day postoperative CSF leaks (p = 0.46). In particular, postradiation CSF leaks occurred in four patients, of whom only two had smoking history, but this number is too small to draw statistical conclusions.

It is important to remember that our RT-related complications include not only wound healing complications but also infectious diseases. Particularly in the sinonasal mucosa, smoking negatively impacts mucociliary clearance and the inflammatory response.[36] These are the primary bactericidal mechanisms for elimination of pathogens[37] and radiation produces similar damage to smoking.[24] In this scenario and according to our results, smoking history might be an additive factor to the inherent risk of RT in producing sinonasal infectious diseases. Nevertheless, it is worth mentioning that perioperative abstinence from smoking might improve the inflammatory response in the wound but not the remodeling and healing process.[13] [38]

The use of neoadjuvant chemotherapy has been studied as a potential risk factor for perioperative morbidity and mortality, but results are inconsistent. Some authors reported a slight increase in the perioperative risk[39] [40] [41] and others found no association between neoadjuvant chemotherapy and early postoperative complications, including surgical wound infections.[42] [43] In our study, almost every patient who was treated with neoadjuvant chemotherapy had sinonasal epithelial malignancies and infectious diseases; this group represented more than two-thirds of the post-RT complications (73%). The use of chemotherapy is an important risk factor for neutropenia and associated complications, and most of these events occur within the first two cycles.[44] Particularly for sinonasal cancers, taxane-based and/or platinum-based drug combinations are frequently used[45] [46] and these combinations are among the ones with higher risk of neutropenia and myelosuppression.[47] [48] Curiously, Turan et al found that the odds of 30-day postoperative morbidities (including surgical wound infection) did not differ between patients who were or were not given chemotherapy within 30 days before the surgery.[42] Similar results were reported by Fahy et al in surgical resection of liver metastasis.[49] However, in these studies, chemotherapy regimens were not stratified, and a longer safe period might be warranted for specific drugs. There is no clear consensus regarding a safe time interval between ending chemotherapy and undergoing an elective surgical intervention, but most authors agree that an interval ≥4 weeks has a lower perioperative risk,[50] and the longer the time between the completion of chemotherapy and the elective surgery, the lower the complication rates.[51] Nevertheless, after a multivariate analysis, we did not find a significant increase in risk of complications with the use of chemotherapy. These ambiguous results could explain the lack of significance of neoadjuvant chemotherapy in the multivariate analysis. To our knowledge, our study is the first to analyze the impact of neoadjuvant chemotherapy as an independent risk factor for postradiation complications in sinonasal malignancies. Specific chemotherapy agents and the time to elective surgery were beyond the objective of our analysis but could represent a starting point for future studies.

Finally, the type of radiation and radiation fields used was another factor that modified the risk of RT-related complications. Given the high doses of radiation required for tumor treatment, reducing the dose to adjacent normal tissues has been a main focus in the treatments of head and neck cancers. Modern photon therapy including intensity-modulated RT has offered a better quality of life for these patients.[52] More recently, intensity-modulated proton therapy has allowed further reductions in dose to surrounding normal tissues resulting in an even greater reduction in toxicities.[53] The capacity of proton beams to deposit the bulk of their radiation dose in a highly confined area has been particularly beneficial for deep-seated tumors, such as those in the mid-skull base, and for those tumors close to other major organs, like head and neck cancers.[54] As mentioned earlier, the subepithelial edema of the mucosa (mucositis) is a well-known complication of RT, but this mucositis is significantly lower with the use of proton-based therapies.[55] This could explain our finding that RT-related complications are more frequent in patients receiving photon-based radiation, but this finding is not an independent predictor of RT complications. After the multivariate analysis, this association was not significant, and this might be related to the type of tumor and the radiation field used for each technique. Given the potential for mucosal and submucosal spread, sinonasal malignancies require larger clinical target volumes to include high-risk regions such as the adjacent sinuses.[56] [57] Because of these factors, and limited availability of proton therapy, photon-based therapies are still a mainstay of treatment for sinonasal malignancies. Including the sinuses in the radiation field of photon-based therapies might explain the increased risk of RT-related complications in those patients.

To conclude, while our study has several limitations, our findings may provide guidance for prospective studies with regard to delays in the initiation of adjuvant RT. First limitation is this is a single-institution retrospective study, and our data are subject to selection bias, recall bias, and heterogeneity. Second, we included in our analysis all the patients treated with radiation without randomization and there is no control group, which limits our ability to draw conclusions about which patients might benefit with a delay in their adjuvant radiation. Third, we were limited to complications data that were present in our records for these patients. It is possible that some patients sought care for post-RT complications at other medical facilities, and thus these data not documented in our records. Fourth, the sizes of the bony and dural defects were not analyzed quantitatively; future studies could consider these variables separately. Finally, surgical techniques differ among surgeons and might be different for each tumor, which may affect the risk of complications.

Conclusion

We present the results of a retrospective study analyzing the time to initiation of RT after endoscopic endonasal surgery. We concluded that there is no increased risk of RT-related complications in those patients receiving radiation within 6 weeks after surgery. While early RT might not affect the wound healing process, special attention must be placed in patients who are smokers, who were treated with induction chemotherapy, and those with sinonasal malignancies undergoing photon-based RT.

Conflict of Interest

S.M.R. serves as a consultant to Stryker.

• References

• 1 Larnaudie A, Delaby N, Marcy PY. et al. Radiotherapy of sinonasal cancers. Cancer Radiother 2022; 26 (1–2): 156-167
• 2 Huang J, Barbera L, Brouwers M, Browman G, Mackillop WJ. Does delay in starting treatment affect the outcomes of radiotherapy? A systematic review. J Clin Oncol 2003; 21 (03) 555-563
• 3 Mackillop WJ. Killing time: the consequences of delays in radiotherapy. Radiother Oncol 2007; 84 (01) 1-4
• 4 Mackillop WJ, Bates JH, O'Sullivan B, Withers HR. The effect of delay in treatment on local control by radiotherapy. Int J Radiat Oncol Biol Phys 1996; 34 (01) 243-250
• 5 Harris JP, Chen MM, Orosco RK, Sirjani D, Divi V, Hara W. Association of survival with shorter time to radiation therapy after surgery for US patients with head and neck cancer. JAMA Otolaryngol Head Neck Surg 2018; 144 (04) 349-359
• 6 Dormand EL, Banwell PE, Goodacre TEE, Goodacre TE. Radiotherapy and wound healing. Int Wound J 2005; 2 (02) 112-127
• 7 Devalia HL, Mansfield L. Radiotherapy and wound healing. Int Wound J 2008; 5 (01) 40-44
• 8 Gieringer M, Gosepath J, Naim R. Radiotherapy and wound healing: principles, management and prospects (review). Oncol Rep 2011; 26 (02) 299-307
• 9 Huang WH, Liu CM, Chao TK, Hung PK. Middle meatus bacteriology of acute rhinosinusitis in patients after irradiation of nasopharynx. Am J Rhinol 2007; 21 (03) 286-288
• 10 Surico G, Muggeo P, Mappa L. et al. Impairment of nasal mucociliary clearance after radiotherapy for childhood head cancer. Head Neck 2001; 23 (06) 461-466
• 11 Xiao R, Joshi RR, Husain Q. et al. Timing of surgery and adjuvant radiation therapy for sinonasal malignancies: effect of surgical approach. Head Neck 2019; 41 (10) 3551-3563
• 12 Guo S, Dipietro LA. Factors affecting wound healing. J Dent Res 2010; 89 (03) 219-229
• 13 Sorensen LT, Karlsmark T, Gottrup F. Abstinence from smoking reduces incisional wound infection: a randomized controlled trial. Ann Surg 2003; 238 (01) 1-5
• 14 Wilson JA, Clark JJ. Obesity: impediment to postsurgical wound healing. Adv Skin Wound Care 2004; 17 (08) 426-435
• 15 Kassam AB, Prevedello DM, Carrau RL. et al. Endoscopic endonasal skull base surgery: analysis of complications in the authors' initial 800 patients. J Neurosurg 2011; 114 (06) 1544-1568
• 16 Dehdashti AR, Stofko D, Okun J, Obourn C, Kennedy T. Endoscopic endonasal reconstruction of skull base: repair protocol. J Neurol Surg B Skull Base 2016; 77 (03) 271-278
• 17 Bütof R, Baumann M. Time in radiation oncology - keep it short!. Radiother Oncol 2013; 106 (03) 271-275
• 18 Zhang M, Xu F, Ni W. et al. Survival impact of delaying postoperative chemoradiotherapy in newly-diagnosed glioblastoma patients. Transl Cancer Res 2020; 9 (09) 5450-5458
• 19 Bese NS, Sut PA, Ober A. The effect of treatment interruptions in the postoperative irradiation of breast cancer. Oncology 2005; 69 (03) 214-223
• 20 Thames HD, Kuban D, Levy LB. et al. The role of overall treatment time in the outcome of radiotherapy of prostate cancer: an analysis of biochemical failure in 4839 men treated between 1987 and 1995. Radiother Oncol 2010; 96 (01) 6-12
• 21 Jang JW, Chan AW. Prevention and management of complications after radiotherapy for skull base tumors: a multidisciplinary approach. Adv Otorhinolaryngol 2013; 74: 163-173
• 22 Peters LJ, Goepfert H, Ang KK. et al. Evaluation of the dose for postoperative radiation therapy of head and neck cancer: first report of a prospective randomized trial. Int J Radiat Oncol Biol Phys 1993; 26 (01) 3-11
• 23 Adelstein D, Gillison ML, Pfister DG. et al. NCCN guidelines® insights head and neck cancers, version 2.2017 featured updates to the NCCN guidelines. J Natl Compr Canc Netw 2017; 15 (06) 761-770
• 24 Kuhar HN, Tajudeen BA, Heilingoetter A. et al. Distinct histopathologic features of radiation-induced chronic sinusitis. Int Forum Allergy Rhinol 2017; 7 (10) 990-998
• 25 Hsin CH, Tseng HC, Lin HP, Chen TH. Post-irradiation otitis media, rhinosinusitis, and their interrelationship in nasopharyngeal carcinoma patients treated by IMRT. Eur Arch Otorhinolaryngol 2016; 273 (02) 471-477
• 26 Lu YT, Lu YC, Cheng HC. et al. Chronic rhinosinusitis after radiotherapy in patients with head and neck cancer: a population-based cohort study in Taiwan. Int Forum Allergy Rhinol 2020; 10 (05) 692-697
• 27 Su YX, Liu LP, Li L. et al. Factors influencing the incidence of sinusitis in nasopharyngeal carcinoma patients after intensity-modulated radiation therapy. Eur Arch Otorhinolaryngol 2014; 271 (12) 3195-3201
• 28 Ku PK, Tong MC, Tsang SS, van Hasselt A. Acquired posterior choanal stenosis and atresia: management of this unusual complication after radiotherapy for nasopharyngeal carcinoma. Am J Otolaryngol 2001; 22 (04) 225-229
• 29 Shepard PM, Houser SM. Choanal stenosis: an unusual late complication of radiation therapy for nasopharyngeal carcinoma. Am J Rhinol 2005; 19 (01) 105-108
• 30 Kamel R, Al-Badawy S, Khairy A, Kandil T, Sabry A. Nasal and paranasal sinus changes after radiotherapy for nasopharyngeal carcinoma. Acta Otolaryngol 2004; 124 (04) 532-535
• 31 Wu PW, Huang CC, Lee YS. et al. Post-irradiation sinus mucosa disease in nasopharyngeal carcinoma patients treated with intensity-modulated proton therapy. Cancers (Basel) 2022; 14 (01) 225
• 32 Crippen MM, Patel N, Filimonov A. et al. Association of smoking tobacco with complications in head and neck microvascular reconstructive surgery. JAMA Facial Plast Surg 2019; 21 (01) 20-26
• 33 Silverstein P. Smoking and wound healing. Am J Med 1992; 93 (1A): 22S-24S
• 34 Torres-Bayona S, Velasquez N, Nakassa A. et al. Risk factors and reconstruction techniques for persistent cerebrospinal fluid leak in patients undergoing endoscopic endonasal approach to the posterior fossa. J Neurol Surg B Skull Base 2021; 83 (Suppl. 02) e318-e323
• 35 Thorp BD, Sreenath SB, Ebert CS, Zanation AM. Endoscopic skull base reconstruction: a review and clinical case series of 152 vascularized flaps used for surgical skull base defects in the setting of intraoperative cerebrospinal fluid leak. Neurosurg Focus 2014; 37 (04) E4
• 36 Hutson K, Clark A, Hopkins C. et al; CRES Group. Evaluation of smoking as a modifying factor in chronic rhinosinusitis. JAMA Otolaryngol Head Neck Surg 2021; 147 (02) 159-165
• 37 Gottrup F. Oxygen in wound healing and infection. World J Surg 2004; 28 (03) 312-315
• 38 Sørensen LT, Toft B, Rygaard J, Ladelund S, Teisner B, Gottrup F. Smoking attenuates wound inflammation and proliferation while smoking cessation restores inflammation but not proliferation. Wound Repair Regen 2010; 18 (02) 186-192
• 39 Pisters KM, Vallières E, Crowley JJ. et al. Surgery with or without preoperative paclitaxel and carboplatin in early-stage non-small-cell lung cancer: Southwest Oncology Group Trial S9900, an intergroup, randomized, phase III trial. J Clin Oncol 2010; 28 (11) 1843-1849
• 40 Nordlinger B, Sorbye H, Glimelius B. et al; EORTC Gastro-Intestinal Tract Cancer Group, Cancer Research UK, Arbeitsgruppe Lebermetastasen und-tumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie (ALM-CAO), Australasian Gastro-Intestinal Trials Group (AGITG), Fédération Francophone de Cancérologie Digestive (FFCD). Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008; 371 (9617): 1007-1016
• 41 Kaklamanos IG, Walker GR, Ferry K, Franceschi D, Livingstone AS. Neoadjuvant treatment for resectable cancer of the esophagus and the gastroesophageal junction: a meta-analysis of randomized clinical trials. Ann Surg Oncol 2003; 10 (07) 754-761
• 42 Turan A, Shao D, Salmasi V. et al. Chemotherapy within 30 days before surgery does not augment postoperative mortality and morbidity. Can J Anaesth 2012; 59 (08) 758-765
• 43 Meric F, Milas M, Hunt KK. et al. Impact of neoadjuvant chemotherapy on postoperative morbidity in soft tissue sarcomas. J Clin Oncol 2000; 18 (19) 3378-3383
• 44 Crawford J, Dale DC, Kuderer NM. et al. Risk and timing of neutropenic events in adult cancer patients receiving chemotherapy: the results of a prospective nationwide study of oncology practice. J Natl Compr Canc Netw 2008; 6 (02) 109-118
• 45 Okano S, Tahara M, Zenda S. et al. Induction chemotherapy with docetaxel, cisplatin and S-1 followed by proton beam therapy concurrent with cisplatin in patients with T4b nasal and sinonasal malignancies. Jpn J Clin Oncol 2012; 42 (08) 691-696
• 46 Prestwich RJ, Öksüz DÇ, Dyker K, Coyle C, Şen M. Feasibility and efficacy of induction docetaxel, cisplatin, and 5-fluorouracil chemotherapy combined with cisplatin concurrent chemoradiotherapy for nonmetastatic Stage IV head-and-neck squamous cell carcinomas. Int J Radiat Oncol Biol Phys 2011; 81 (04) e237-e243
• 47 Tian H, Ma D, Tan X. et al. Platinum and taxane based adjuvant and neoadjuvant chemotherapy in early triple-negative breast cancer: a narrative review. Front Pharmacol 2021; 12: 770663
• 48 Lee CK, Gurney H, Brown C. et al. Carboplatin-paclitaxel-induced leukopenia and neuropathy predict progression-free survival in recurrent ovarian cancer. Br J Cancer 2011; 105 (03) 360-365
• 49 Fahy BN, Aloia TA, Jones SL, Bass BL, Fischer CP. Chemotherapy within 30 days prior to liver resection does not increase postoperative morbidity or mortality. HPB (Oxford) 2009; 11 (08) 645-655
• 50 Sullivan MC, Roman SA, Sosa JA. Does chemotherapy prior to emergency surgery affect patient outcomes? Examination of 1912 patients. Ann Surg Oncol 2012; 19 (01) 11-18
• 51 Welsh FKS, Tilney HS, Tekkis PP, John TG, Rees M. Safe liver resection following chemotherapy for colorectal metastases is a matter of timing. Br J Cancer 2007; 96 (07) 1037-1042
• 52 Kam MKM, Leung S-F, Zee B. et al. Prospective randomized study of intensity-modulated radiotherapy on salivary gland function in early-stage nasopharyngeal carcinoma patients. J Clin Oncol 2007; 25 (31) 4873-4879
• 53 Holliday EB, Kocak-Uzel E, Feng L. et al. Dosimetric advantages of intensity-modulated proton therapy for oropharyngeal cancer compared with intensity-modulated radiation: a case-matched control analysis. Med Dosim 2016; 41 (03) 189-194
• 54 Leeman JE, Romesser PB, Zhou Y. et al. Proton therapy for head and neck cancer: expanding the therapeutic window. Lancet Oncol 2017; 18 (05) e254-e265
• 55 Romesser PB, Cahlon O, Scher E. et al. Proton beam radiation therapy results in significantly reduced toxicity compared with intensity-modulated radiation therapy for head and neck tumors that require ipsilateral radiation. Radiother Oncol 2016; 118 (02) 286-292
• 56 Lydiatt WM, Patel SG, O'Sullivan B. et al. Head and neck cancers-major changes in the American Joint Committee on cancer eighth edition cancer staging manual. CA Cancer J Clin 2017; 67 (02) 122-137
• 57 Thawani R, Kim MS, Arastu A. et al. The contemporary management of cancers of the sinonasal tract in adults. CA Cancer J Clin 2023; 73 (01) 72-112

Publication History

Received: 08 March 2023

Accepted: 19 June 2023

Accepted Manuscript online:
21 June 2023

Article published online:
28 July 2023