Introduction
Concurrent chemoradiation is widely used in the definitive treatment of locally advanced head and neck squamous cell carcinoma. Meta-analyses and randomised trials have shown that chemotherapy provides its greatest survival benefit when administered concurrently with radiotherapy [1][2]. Intensity-modulated radiotherapy (IMRT) allows conformal delivery of radiation to complex target volumes while limiting the dose received by nearby normal tissues. Compared with conventional radiotherapy, parotid-sparing IMRT reduces xerostomia and improves recovery of salivary function [3].
Radiotherapy target volumes are defined according to International Commission on Radiation Units and Measurements recommendations [4][5]. The gross tumour volume represents the clinically or radiologically demonstrable disease. The clinical target volume includes areas at risk of microscopic disease, while the planning target volume accounts for geometrical uncertainties such as patient positioning and internal movement. The margins used for clinical and planning target volumes should be based on the primary site, extent of disease, immobilisation method and institutional protocol rather than applying the same margin to every patient.
IMRT may be delivered using a simultaneous integrated boost technique, in which different dose levels are given to the gross disease, high-risk subclinical regions and elective nodal regions during the same treatment fraction. SIB-IMRT permits treatment of these volumes within a single plan and is an established approach for definitive treatment of head and neck cancer [6].
The accuracy of IMRT depends on the relationship between the target volumes and surrounding normal structures remaining reasonably close to that recorded on the planning CT. However, patients with head and neck cancer may develop tumour or nodal shrinkage, weight loss, reduction in oedema and changes in external contour during treatment. These changes may alter the position and volume of both the target and organs at risk [7][9]. Image guidance can correct daily set-up errors but cannot fully compensate for progressive internal anatomical changes when the original treatment plan is retained.
Adaptive radiotherapy involves repeating imaging and modifying the treatment plan when anatomical or dosimetric changes are considered clinically important [10]. Studies using repeat CT have shown that failure to account for these changes may reduce target coverage or increase the dose received by the spinal cord, brainstem and salivary glands [8][11][12]. Adaptive replanning may therefore help maintain the intended dose distribution during treatment.
The need for replanning is not the same in every patient. Tumour site and stage, initial tumour volume, weight loss, changes in external contour and reduction in tumour or nodal volume may influence the need for adaptation [13][15]. Prospective studies have reported improvements in target coverage and reductions in selected organ-at-risk doses following replanning [16][17]. However, most published studies are small, non-randomised and clinically heterogeneous, and the reported dosimetric improvements have not consistently translated into better toxicity or disease-control outcomes [18][23].
The timing and frequency of replanning also remain uncertain. Some centres use a scheduled repeat CT during the second half of treatment, while others use changes detected during image guidance, weight loss or mask fit to trigger reassessment [19][20][24]. Patient-reported symptoms and MR-guided treatment have also been investigated as possible methods for selecting patients who require adaptation [25][26]. A scheduled single replan may be more practical in centres where the resources required for repeated imaging, contouring and planning are limited.
Recent prospective studies have continued to examine clinical outcomes, resource requirements and patient-selection criteria for adaptive radiotherapy [27][28]. A systematic review found that adaptive radiotherapy produced small reductions in doses to organs at risk, but its effect on toxicity and disease control remained uncertain [29]. Similarly, a phase III randomised trial of weekly adaptive radiotherapy in patients with oropharyngeal cancer did not show a significant reduction in xerostomia compared with standard IMRT, although improvement in parotid excretory function was observed [30]. These findings support a selective and carefully assessed approach rather than routine repeated replanning for every patient.
Replanning is particularly relevant for IMRT, which is characterized by high conformality to the tumour volume and steep dose gradients around normal tissue, the accuracy of such a plan can be compromised by alterations in patient anatomy and shrinkage of the target volume during treatment. Treatment plans optimized on a single pre-treatment CT can become suboptimal for later fractions as the underlying structures deform and dose-volume histograms based on that single scan may then misestimate the dose actually delivered. SIB-IMRT and adaptive radiotherapy are established techniques. This study reports the extent and patient-to-patient variation in primary tumour-volume reduction at a planned reassessment at 40 Gy. It also describes our institutional experience of including a single scheduled adaptive replanning procedure in routine treatment.
The aim of this study is to evaluate the change in primary gross tumour volume at 40 Gy and describe the feasibility and acute toxicity of planned adaptive replanning during definitive SIB-IMRT for locally advanced head and neck cancers.
Materials & Methods
Study design and setting: This retrospective observational study reviewed the records of patients with carcinoma of the oral cavity, oropharynx or nasopharynx who received definitive chemoradiation between January 2018 and August 2019.
Primary endpoint: Absolute and percentage change in primary GTV between baseline planning CT and replanning CT at 40 Gy.
Secondary endpoints: Completion of replanning and treatment, treatment interruptions, and acute dermatitis and mucositis.
Study Participants: Fifteen patients who had been treated with SIB-IMRT and who had undergone a replanning CT at a tumour dose of 40 Gy were identified for inclusion in this analysis.
Inclusion criteria: Patients were included if they had primary carcinoma of the oral cavity, oropharynx or nasopharynx; AJCC stage III or IV disease; an ECOG performance status of 0 or 1; and received definitive SIB-IMRT with three-weekly cisplatin. Both baseline and repeat planning CT scans obtained at around 40 Gy.
Exclusion criteria: If patients had received prior head and neck irradiation, had an ECOG performance status more than or equal to 2, received treatment with palliative intent, presented with recurrent or distant metastatic disease at baseline, were medically unfit for concurrent systemic therapy or underwent upfront surgical resection.
Study Procedure: The organs at risk contoured and evaluated during treatment optimization included the bilateral parotid glands, bilateral submandibular glands, spinal cord, brainstem, pharyngeal constrictor muscles, mandible, optic chiasm, optic nerves, retinas, ocular lenses, cochleae, pituitary gland, and hippocampi.
Patients were immobilized using a perforated thermoplastic head and neck mask indexed to the couch with a patient positioning system to ensure reproducible set up. Lead markers were placed on relevant anatomical bony landmarks. Contrast enhanced planning CT scans were obtained at a 2.5 mm slice thickness. Baseline investigations included a hemogram, renal and liver function tests, serum electrolytes, coagulation profile, fasting and post-prandial blood glucose, blood grouping, viral markers, chest X-ray, echocardiogram, and direct laryngoscopy/pharyngoscopy with biopsy of the lesion, weekly investigations during treatment included a hemogram, serum electrolytes and renal function tests.
Radiotherapy was delivered once daily, five days a week over 33 fractions. The GTV was defined as the clinical and radiological extent of tumour, the CTV as a 0.5-1 cm margin around the GTV to account for microscopic disease, and the PTV as a further 0.3-0.5 cm margin around the CTV to account for organ motion and set-up error with all volumes contoured per ICRU 62 guidelines. The high-risk CTV received a total dose of 66-70 Gy (2.12-2.20 Gy/fraction), the intermediate-risk CTV 60 Gy (2.0 Gy/fraction) and the low-risk CTV 54 Gy (1.8 Gy/fraction) all delivered simultaneously using the SIB technique. Tolerance doses to the eyes, lens, brain stem, spinal cord and parotids were not exceeded. Cone-beam CT was acquired daily to verify set-up and a replanning CT was obtained at a tumour dose of 40 Gy to assess volumetric change, re-contouring at replanning followed RTOG contouring guidelines and all planning was performed on the Eclipse treatment planning system.
All patients received concurrent 3-weekly cisplatin (100 mg/m² intravenously with standard antiemetic premedication and pre- and post-hydration).
As per the institutional protocol, repeat planning CT was performed at around 40 Gy. This allowed assessment of anatomical changes after a major part of the prescribed dose had been delivered, while leaving enough treatment fractions for use of the revised plan. The radiation oncologist reviewed the clinical and radiological response, recontoured the target volumes and organs at risk, checked the revised dose distribution and approved the adaptive plan.
Patients were assessed weekly during chemoradiation for treatment tolerance, acute toxicity and clinical changes in the tumour. Tumour regression was evaluated by comparing the primary gross tumour volume on the baseline planning CT with that on the repeat planning CT obtained at 40 Gy. Acute and late radiation toxicities were graded using the RTOG morbidity scoring criteria. The first follow-up assessment was carried out six weeks after treatment, followed by monthly visits during the first year, once every three months during the second year and once every six months thereafter. The present analysis included follow-up data from the first year only. Follow-up included physical and oral cavity examination, with indirect or direct laryngoscopy when required. Further imaging was performed when clinically indicated. [8].
Study Analysis: The objectives of this analysis were to evaluate the change in primary gross tumour volume at 40 Gy and describe the feasibility and acute toxicity of planned adaptive replanning during definitive SIB-IMRT for locally advanced head and neck cancers. Table 1 shows the baseline demographic, disease and treatment characteristics of the study population.
Continuous variables were tested for normality using the Shapiro–Wilk test. As the paired differences in primary tumour volume were not normally distributed, the baseline and 40-Gy volumes were compared using the two-sided Wilcoxon signed-rank test. The mean absolute reduction was reported with a 95% confidence interval. Percentage reduction was expressed as median and range. Categorical variables and treatment-related toxicities were presented as numbers and percentages. A p value of <0.05 was considered statistically significant. Data were analysed using SPSS version 25.0. The 95% confidence interval for the mean absolute reduction was calculated using the t distribution.
Results
Baseline demographic, disease and treatment characteristics are presented in Table 1. Of the 15 patients, 11 (73.33%) were male and 4 (26.67%) were female. Most patients were aged between 30 and 49 years (10/15, 66.67%). Seven patients (46.67%) had T3 disease and eight (53.33%) had T4 disease. The definitions of the volumetric parameters are provided in Table 2. Individual patient volume metrics are summarized in Table 3.
Site-wise tumour-volume response at 40 Gy is given in Table 4. The median primary gross tumour volume decreased from 132.41 cm³ at baseline to 104.49 cm³ at 40 Gy. The reduction was statistically significant on Wilcoxon signed-rank testing (W = 0, p < 0.001). Site-wise findings are descriptive because the small and unequal subgroups do not permit reliable statistical comparison.
| Characteristic | Category | Number of patients, n (%) |
| Primary tumour site | Oral cavity | 7 (46.67) |
| Oropharynx | 6 (40.00) | |
| Nasopharynx | 2 (13.33) | |
| Sex | Male | 11 (73.33) |
| Female | 4 (26.67) | |
| Age group (years) | 20–29 | 3 (20.00) |
| 30–39 | 5 (33.33) | |
| 40–49 | 5 (33.33) | |
| 50–59 | 2 (13.33) | |
| T category (AJCC Cancer Staging Manual, 8th Edition) | T3 | 7 (46.67) |
| T4 | 8 (53.33) | |
| Concurrent chemotherapy completion | Completed three cycles of three-weekly cisplatin | 15 (100.00) |
| Parameter | Definition |
| TV | Volume of the primary tumour measured before RT |
| TV-40 | Volume of the residual primary tumour measured after 40 Gy of radiotherapy |
| TRR-40 | (TV − TV-40) / TV × 100 |
| TR rate-40 | TRR-40 / number of days between the first and second CT scan |
TV - Tumour Volume; TRR – Tumour Regression Ratio
| Patient (Site) | TV (cm³) | TV-40 (cm³) | TRR-40 (%) | TR rate-40 (%/day) |
| 1 (Oral cavity) | 50.75 | 36.07 | 28.92 | 1.40 |
| 2(Oral cavity) | 44.19 | 36.65 | 17.06 | 0.80 |
| 3(Oral cavity) | 41.87 | 31.05 | 25.84 | 1.36 |
| 4(Oral cavity) | 84.45 | 82.25 | 2.60 | 0.13 |
| 5(Oral cavity) | 74.13 | 40.77 | 45.00 | 2.14 |
| 6(Oral cavity) | 352.60 | 299.16 | 15.81 | 0.79 |
| 7(Oral cavity) | 248.27 | 141.87 | 42.78 | 2.13 |
| 8(Oropharynx) | 56.28 | 44.70 | 20.57 | 1.02 |
| 9(Oropharynx) | 94.00 | 69.68 | 25.87 | 1.36 |
| 10(Oropharynx) | 196.65 | 155.44 | 20.95 | 0.95 |
| 11(Oropharynx) | 132.41 | 104.49 | 21.08 | 1.05 |
| 12(Oropharynx) | 327.75 | 300.72 | 8.24 | 0.41 |
| 13(Oropharynx) | 266.25 | 222.50 | 16.43 | 0.82 |
| 14(Nasopharynx) | 176.25 | 149.27 | 15.30 | 0.80 |
| 15(Nasopharynx) | 256.55 | 215.77 | 15.89 | 0.83 |
Baseline tumour volume (TV), residual tumour volume at 40 Gy (TV-40), tumour regression ratio (TRR-40), and tumour regression rate (TR rate-40) across individual patients.
| Primary site | n | Baseline TV (cm³), mean (SD) | TV-40 (cm³), mean (SD) | Absolute reduction (cm³), mean (SD) | TRR-40 (%), mean (SD) | TR rate-40 (%/day), mean (SD) |
| Oral cavity | 7 | 128.04 (122.56) | 95.40 (98.27) | 32.63 (37.03) | 25.43 (15.17) | 1.25 (0.74) |
| Oropharynx | 6 | 178.89 (104.41) | 149.59 (97.58) | 29.30 (11.80) | 18.86 (6.00) | 0.94 (0.31) |
| Nasopharynx | 2 | 216.40 (56.78) | 182.52 (47.02) | 33.88 (9.76) | 15.60 (0.42) | 0.82 (0.02) |
| Overall | 15 | 160.16 (108.07) | 128.69 (94.09) | 31.47 (25.45) | 21.49 (11.28) | 1.07 (0.55) |
TV: baseline primary gross tumour volume; TV-40: primary gross tumour volume at 40 Gy; TRR-40: tumour regression ratio at 40 Gy; TR rate-40: tumour regression rate at 40 Gy. The reduction was statistically significant on Wilcoxon signed-rank testing (W = 0, p < 0.001). Site-wise findings are descriptive because the small and unequal subgroups do not permit reliable statistical comparison.
Table 5 shows the treatment completion and acute radiation-related toxicities data. All 15 patients completed the planned chemoradiation course, including three cycles of three-weekly cisplatin. Grade 2 and grade 3 acute radiation dermatitis occurred in 11 (73.33%) and 4 (26.67%) patients, respectively. Grade 2 oral mucositis was observed in 12 patients (80.00%), while three patients (20.00%) developed grade 3 mucositis. All grade 3 toxicities were managed with supportive care on an outpatient basis. No chemotherapy dose was reduced or omitted, and no treatment interruption occurred because of toxicity. As there was no comparison group, the toxicity observed in this study could not be compared with that of other treatment approaches.
| Clinical outcome | Category | n (%) |
| Completion of planned chemoradiation | Completed | 15 (100.00) |
| Completion of three cycles of three-weekly cisplatin | Completed | 15 (100.00) |
| Acute radiation dermatitis | Grade 2 | 11 (73.33) |
| Grade 3 | 4 (26.67) | |
| Acute oral mucositis | Grade 2 | 12 (80.00) |
| Grade 3 | 3 (20.00) | |
| Unplanned treatment interruption due to toxicity | Present | 0 (0.00) |
| Absent | 15 (100.00) |
The mean absolute reduction in primary gross tumour volume at 40 Gy was 31.47 cm³ (95% CI: 17.37–45.56 cm³). The median percentage reduction was 20.57%, with observed reductions ranging from 2.60% to 45.00%. Eight patients (53.33%) demonstrated a reduction greater than 20%. [Table 6] These findings indicate substantial inter-patient variation in tumour regression during treatment and support reassessment of tumour and normal-tissue anatomy at approximately 40 Gy. However, volumetric regression alone does not establish a mandatory need for adaptive replanning; the decision should be individualised according to anatomical changes and their dosimetric effect on target coverage and organs at risk.
| Tumour-response parameter | Result |
| Mean absolute volume reduction, cm³ (95% CI) | 31.47 (17.37–45.56) |
| Median percentage reduction | 20.57% |
| Percentage-reduction range | 2.60–45.00% |
| TRR-40 >20% | 8 (53.33%) |
| TRR-40 ≤20% | 7 (46.67%) |
Discussion
Patients receiving fractionated radiotherapy for head and neck cancer undergo marked anatomic changes during treatment including shrinkage of the primary tumour or nodal mass, resolution of postoperative changes such as oedema and changes in body habitus from weight loss [31][33]. These changes become measurable and clinically significant chiefly in the second half of treatment (after the third week) and can have a dosimetric impact when highly conformal techniques are used [31][32][34]. In one study by Diaz A et al, they found significant difference in estimating dose delivered to the right parotid gland using final week CBCT between using rigid and deformable registration techniques [35]. The SIB-IMRT technique which shortens overall treatment time and increases the dose per fraction to the boost volume represents a way to investigate accelerated fractionation in the definitive treatment of head and neck cancer [36]. Adaptive replanning may be useful when changes occur in the tumour and surrounding normal tissues during treatment [24,37]. However, comparative dose-volume data were not available, and its effect on target coverage and doses to organs at risk could not be evaluated.
The observed reduction in primary tumour volume demonstrates measurable anatomical change during treatment [31][33]. However, the study was not designed or adequately powered to determine whether adaptive replanning or SIB-IMRT improved local control or survival. Accelerated repopulation can compromise tumour control after conventional radiotherapy for fast-growing tumours [39]. The higher dose per fraction delivered throughout treatment also helps to decrease overall treatment time and hence counter accelerated repopulation [36][40]. Although altered fractionation may have radiobiological advantages, the present study did not evaluate local control or survival. Therefore, no conclusion regarding the oncological efficacy of SIB-IMRT or adaptive replanning can be drawn from these findings.
Clinical and dosimetric studies confirm that mid-course replanning counteracts unintended dose shifts [24,37]. Castelli et al. demonstrated that unadapted IMRT led to an unintended mean parotid overdose of 3.7 Gy in 59% of evaluated glands, whereas adaptive replanning recovered parotid sparing and decreased projected xerostomia risk by 11% (p<0.01) [41]. Similarly, Schwartz et al. observed in a prospective adaptive trial that replanning at mid-treatment significantly lowered the mean dose to both ipsilateral (p=0.002) and contralateral (p=0.003) parotids while preserving high locoregional control rates [17].
The study included cancers of the oral cavity, oropharynx and nasopharynx, resulting in a clinically heterogeneous group. As the number of patients in each subgroup was small and unequal, the differences in tumour regression between these sites could not be reliably compared [29,37]. Grade 3 oral mucositis and dermatitis occurred in 20.0% and 26.7% of patients, respectively. All patients completed the planned chemoradiation without toxicity-related treatment interruption. However, because the study did not include a comparison group, no conclusion can be made regarding the relative toxicity of this treatment approach [30].
When the gross tumour volume is modified during replanning, the clinical and radiological response should be reviewed carefully to avoid a geographical miss. Changes in the patient’s external contour, target position and surrounding organs at risk should also be considered before deciding to replan [24,37,45].
SIB-IMRT and adaptive radiotherapy are established techniques. This study reports the extent and patient-to-patient variation in primary tumour-volume reduction at a planned reassessment at 40 Gy. It also describes our institutional experience of including a single scheduled adaptive replanning procedure in routine treatment.
This analysis is limited by its retrospective design, small sample size (n = 15), single-institution setting and one-year follow-up period. The inclusion of only patients who completed chemoradiation and underwent repeat planning CT may have introduced selection bias and may overestimate the feasibility and treatment-completion rate.
Conclusion
In this retrospective study of 15 patients with locally advanced head and neck cancer, repeat planning CT performed at around 40 Gy showed a significant reduction in primary gross tumour volume, although the degree of reduction varied between patients. All patients completed the planned chemoradiation, and adaptive replanning at this stage was feasible in our centre. However, the small and heterogeneous study group, retrospective design, lack of a comparison group and absence of comparative dosimetric and survival data limit the interpretation of these findings. Larger prospective studies are needed to identify patients who are most likely to benefit from adaptive replanning and to determine the most appropriate time for replanning.
Declarations
Conflict of interest
None
Funding/ financial support
None
Contributors
Dr. Vengada Krishnan P.R, M.D, Department of Radiation Oncology, Cancer Institute [WIA], The Tamil Nadu M.G.R Medical University, Chennai, India.
Dr. Ramanaiah K M.D, Department of Radiation Oncology, Cancer Institute [WIA], The Tamil Nadu M.G.R Medical University, Chennai, India.
Dr. Gopika Rajeev M.D, Department of Radiation Oncology, Cancer Institute [WIA], The Tamil Nadu M.G.R Medical University, Chennai, India.
Ethical Clearance
The study was approved by the Institutional Ethics Committee of Cancer Institute [WIA], The Tamil Nadu M.G. R Medical University, Chennai, India; Reg. No: ECR/235/Inst/TN2013/RR-16: Ref no: IEC/2018/ST14.
Written informed consent for treatment and use of clinical data was obtained at the time of care.
Trial details
Not applicable
Acknowledgements
None