Computed Tomography Number Changes Observed During Computed Tomography–Guided Radiation Therapy for Head and Neck Cancer

International Journal of

Radiation Oncology biology

physics

www.redjournal.org

Physics Contribution

Computed Tomography Number Changes Observed During Computed TomographyeGuided Radiation Therapy for Head and Neck Cancer Mei Feng, MD,*,y Cungeng Yang, PhD,* Xiaojian Chen, PhD,* Shouping Xu, MSc,* Ion Moraru, PhD,* Jinyi Lang, MD,y Christopher Schultz, MD,* and X. Allen Li, PhD* *Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, Wisconsin; and y Department of Radiation Oncology, Sichuan Cancer Hospital, Chengdu, China Received Jun 10, 2014, and in revised form Dec 15, 2014. Accepted for publication Dec 20, 2014.

Summary Radiation-induced CT number changes in gross tumor volume (GTV) and organs at risk in head and neck cancer (HNC) were determined from diagnostic-quality CTs acquired during daily CTguided radiation therapy for HNC. The CT numbers in GTV and parotid glands were reduced for a subset of patients and were correlated with the doses received.

Purpose: To investigate CT number (CTN) changes in gross tumor volume (GTV) and organ at risk (OAR) according to daily diagnostic-quality CT acquired during CTguided intensity modulated radiation therapy for head and neck cancer (HNC) patients. Methods and Materials: Computed tomography scans acquired using a CT-on-rails during daily CT-guided intensity modulated radiation therapy for 15 patients with stage II to IVa squamous cell carcinoma of the head and neck were analyzed. The GTV, parotid glands, spinal cord, and nonspecified tissue were generated on each selected daily CT. The changes in CTN distributions and the mean and mode values were collected. Pearson analysis was used to assess the correlation between the CTN change, organ volume reduction, and delivered radiation dose. Results: Volume and CTN changes for GTV and parotid glands can be observed during radiation therapy delivery for HNC. The mean (SD) CTNs in GTV and ipsi- and contralateral parotid glands were reduced by 6  10, 8  7, and 11  10 Hounsfield units, respectively, for all patients studied. The mean CTN changes in both spinal cord and nonspecified tissue were almost invisible (<2 Hounsfield units). For 2 patients studied, the absolute mean CTN changes in GTV and parotid glands were strongly correlated with the dose delivered (P<.001 and P<.05, respectively). For the correlation between CTN reductions and delivered isodose bins for parotid glands, the Pearson coefficient varied from 0.98 (P<.001) in regions with low-dose bins to 0.96 (P<.001) in high-dose bins and were patient specific. Conclusions: The CTN can be reduced in tumor and parotid glands during the course of radiation therapy for HNC. There was a fair correlation between CTN reduction and radiation doses for a subset of patients, whereas the correlation between CTN

Reprint requests to: X. Allen Li, PhD, Department of Radiation Oncology, Medical College of Wisconsin, 8701 Watertown Plank Rd, Milwaukee, WI 53226. Tel: (414) 805-4362; E-mail: [email protected] Int J Radiation Oncol Biol Phys, Vol. 91, No. 5, pp. 1041e1047, 2015 0360-3016/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijrobp.2014.12.057

This work was supported in part by Siemens Medical Solutions USA, Inc. Conflict of interest: none.

International Journal of Radiation Oncology  Biology  Physics

1042 Feng et al.

reductions and volume reductions in GTV and parotid glands were weak. More studies are needed to understand the mechanism for the radiation-induced CTN changes. Ó 2015 Elsevier Inc. All rights reserved.

Introduction Image guided intensity modulated radiation therapy (IMRT) is the standard treatment for head and neck cancer (HNC) (1). A variety of image-guided technologies have been used to account for interfractional variations, including setup uncertainty and anatomic changes. The anatomic changes, including changes in the volumes and shapes of tumor and organs at risk (OARs), are the major interfractional variations observed during the course of RT delivery for HNC (1, 2). These changes, however, cannot be fully accounted for by the current practice of image guided radiation therapy (IGRT) repositioning. Adaptive RT that is capable of addressing these anatomic changes has been introduced in recent years (3, 4). One of the unsolved issues for adaptive RT of HNC is the timing and/or indicators to trigger adaptive replanning during the course of RT delivery. It has been reported that the CT number (CTN) for tumor and certain normal structures can change after irradiation, and the CTN change may be an early indicator for local control and/or radiation-induced damage (5-8). For example, Diot et al (5) and Palma et al (6) independently reported that normal lung density (ie CTN) was changed after stereotactic body RT (SBRT) for lung cancer, as observed from follow-up CTs, and the CTN changes were related to radiation doses. Howells et al (7) reported that reduced normal liver tissue density was correlated with radiation dose after SBRT for liver cancer. Mayer et al (8) found that the reduction of tumor CTN in lung cancer patients after RT might be associated with local control. In this work we conducted a study to investigate the CTN changes for both tumor and OARs according to daily diagnostic-quality CTs collected during IG-IMRT for patients with HNC. An effort will be made to examine the relationships between CTN changes, volume regressions, and delivered radiation doses during the courses of RT.

Methods and Materials Patient data Diagnostic-quality CTs acquired using a CT-on-rails (CTVision; Siemens Medical Solutions USA, Malvern, PA) during daily CT-guided IMRT for 15 patients with stage II-IVa (staging system, American Joint Committee on Cancer, 6th edition) squamous cell carcinoma of the head and neck were analyzed. The outcome data with a short follow-up time (<1 year) for these patients, treated consecutively, were included in the analysis. Detailed patient characteristics along with various dosimetric

parameters are listed in Table 1. In each radiation fraction, the CT was acquired in the normal acquisition mode (120 kV, image noise 1 Hounsfield unit [HU], slice thickness 3 mm), with the patient immobilized with a thermoplastic mask. The patient was repositioned according to a rigidbody registration of the planning CT and the daily CT before treatment delivery. For all patients, gross tumor volume (GTV) and nodal disease were treated to 70 Gy in 35 fractions. Elective nodal disease was treated to 51 to 63 Gy in 35 fractions with a simultaneously integrated boost technique and concurrently with 2 to 3 cycles of cisplatinbased chemotherapy. For each case studied, daily CT sets for the first and last fractions and 1 fraction per week, a total of 6 CT sets were selected for analysis. The GTV, ipsiand contralateral parotid glands, spinal cord, and a region of nonspecified tissue (NST) (lower-dose region, a part of trapezius below sternum) were generated by populating the corresponding contours from the planning CT to each selected daily CT using an auto-segmentation tool (ABAS, Elekta, Computerized Medical Sysytem, St. Louis, MO), with manual editing if necessary by a radiation oncologist. Every effort was made to minimize the variations in contour generation in this study. The delineation uncertainty should not significantly affect the observed results, because we are only interested in the changes in mean CTNs. A series of isodose lines was also populated from the planning CT to the daily CT in a similar way.

Data analysis An in-house software tool developed with MATLAB (version R2013a; MathWorks, Natick, MA) was used to analyze the selected daily CTs with the generated daily contours of GTV, parotid glands, spinal cord, and NST and generated isodose lines. The CTN (in HU) in each voxel inside each contoured structure was specified using the software tool. In addition, the CTN of each voxel within a region of interest was extracted by Boolean calculation of masks derived from contours. For example, a volume in the contralateral parotid gland with an exposure dose in the range from 20 Gy to 30 Gy over the whole RT course is calculated as follows:

MLeft Parotid AND NOT M30Gy
AND MLeft Parotid And M20Gy ; where AND and NOT are Boolean operators, and MX s are the masks with their self-revealing subscripts. The histograms of the CTNs and their mean and mode values in each structure on selected daily CTs were determined by the in-house software. We considered 400 HU as a

Volume 91  Number 5  2015

CT number change 1043

Table 1 Basic characteristics and dosimetric parameters, including dose to 95% of volume (D95) and mean dose (Dmean), of all patients studied Ipsilateral parotid gland

GTV

Contralateral parotid gland

Spinal cord

NST

Volume D95 Dmean Volume Dmean Volume Dmean D05 Dmean Dmean Patient no. Age (y) Sex Stage Chemotherapy? (cm3) (cGy) (cGy) (cm3) (cGy) (cm3) (cGy) (cGy) (cGy) (cGy) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

61 47 65 58 55 53 55 64 51 56 56 51 70 39 68

M M M M F M M M M F M M M M M

III IVA IVA IVA IVA III IVA IVA III III III III IVA IVA II

Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes No

25.0 28.7 30.8 50.4 36.4 10.8 40.5 22.2 44.3 24.8 32.3 18.6 6.5 10.6 18.9

7070 7348 7615 7229 7251 7235 7231 7234 7154 7242 7111 7099 7078 7174 7083

7367 7567 7898 7477 7495 7444 7391 7430 7427 7490 7261 7255 7262 7527 7286

22.7 38.4 24.7 17.1 25.9 27.7 39.2 36.0 27.3 21.2 28.8 24.2 27.3 12.5 8.9

2978 3894 2399 6693 2848 3094 2785 2561 3438 3179 3274 2544 5576 3560 3134

23.6 31.5 23.1 19.1 20.4 31.3 35.6 38.9 27.6 16.9 25.3 19.7 28.7 11.0 8.1

2812 2717 3002 2580 2950 1978 2443 2334 3344 3212 3340 2357 1806 2984 3281

3570 3969 4153 4122 3943 3850 3895 3146 3494 3997 4175 3789 3857 3886 2999

2249 3050 2987 3511 2785 2941 3178 2560 2461 2794 2816 3267 2496 2240 2507

2503 796 665 1057 729 761 638 669 515 1398 479 2076 2175 3830 947

Abbreviations: F Z female; GTV Z gross tumor volume; M Z male; NST Z nonspecified tissue.

reasonable cutoff point to separate tissue and air in the head and neck region (6) (ie the CTNs below 400 HU were excluded from the data analysis of GTV). Pearson analysis was used to assess the correlations between the CTN change and volume change ratio (ratio of volume change and the volume in the first daily CT set) for GTV and parotid glands and between CTN change and delivered dose (the delivered fraction number). One-way analysis of variance was used to assess the correlation between the local control data and CTN change in GTV. P<.05 was considered statistically significant. To ensure that the CTN change observed was not due to the daily variation of the CT imaging, the CTN stability for the scanner used in this work was studied by acquiring and examining CTs for a known CTN phantom periodically. It was found that there was no noticeable change in CTNs over time, implying that there was no variation in daily CTs that contributed to the CTN changes observed in the HNC patients.

Results For all 15 patients studied, the mean (SD) volume reductions were 18.2  12.8 cm3, 9.9  5.6 cm3, and 9.6  5.4 cm3 for GTVs and ipsi- and contralateral parotid glands, respectively, with mean reduction ratios of 0.67  0.3, 0.42  0.15, and 0.42  0.13 between the first and last radiation fractions. With a few exceptions, the mean CTN of GTVs and parotid glands were reduced from the first to last fractions. The changes in the mean and mode CTN in GTV and ipsiand contralateral parotid glands were 5.9  9.2, 8.5  7.1, and 11.0  10.2 HU and 5.7  6.8, 6.8  10.6, and 11.7  9.2 HU, respectively. The mean CTN reductions were

substantial in 3 of 15 patients (patients 7, 10, and 11) in GTVs, with mean CTN reduced by 10 to 40 HU between the first and last fractions. Similar substantial mean CTN reductions (10-40 HU) were also found in the ipsilateral parotid glands in patients 2, 5, 7, 8, 10, and 14, and in the contralateral parotid glands in patients 2, 4, 7, 8, 10, and 14 (Fig. 1a-c). The changes in the mean and mode CTN values from the first to last fractions were small: 0.2  3.5, 1.6  2.3, and 0.2  4.7, 1.0  5.8 HU, respectively, in spinal cord and NST averaged over the 15 patients (Fig. 1d, e). Figure 2a and b shows the mean CTN reductions in GTV and ipsi- and contralateral parotid glands as a function of the volume reduction ratios from the first to last fractions for the 15 patients studied. It is seen that the correlation between mean CTN reductions and volume reduction ratios were generally weak, with correlation coefficients and P values of rZ0.47 and PZ.13 for GTV, rZ0.21 and PZ.53 for ipsilateral parotid glands, and rZ0.32 and PZ.34 for contralateral parotid glands. For the correlation between CTN change and delivered dose analyzed for all 15 patients, it was observed that the CTN reduction was significantly correlated with the mean dose delivered in patients 7, 8, 10, and 11 for GTV, in patients 2, 4, 8, and 10 for contralateral parotid glands, and in patients 2 and 8 for ipsilateral parotid glands. Among all the 15 patients studied, patient 12 had a biopsy-proven persistent residual tumor, and no radiation-related toxicity was observed. There was no correlation between local control and mean CTN change in GTV (PZ.286). Selected data are presented below. The details of CTN changes during the entire course of treatment for one of the patients (patient 10) exhibiting substantial CTN changes in GTVs are presented in

International Journal of Radiation Oncology  Biology  Physics

1044 Feng et al.

10

0

0

-10

-10

-20

-20

-30

-30

-40

-40 -50

-50

mean CTN reduction mode CTN reduction volume reduction

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Mean CTN reduction (Hu)

A

volume and CT number reduction of GTV 10

volume reduction (cc)

CTN reduction (Hu)

A

Patient number

10

0

0

-10

-10

-20

-20

-30

-30

-40

-40

-50

-50

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

mean CTN reduction mean mode reduction volume reduction

C

0.2

-10

-20

-20

-30

-30

-40

-40 -50

volume reduction (cc)

0

-10

mean CTN reduction mean mode reduction volume reduction

volume reduction (cc)

CTN reduction (Hu)

10

0

mean CTN reduction mean mode reduction volume reduction

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Patient number

D

volume and CTN reduction of spinal cord 10

10

0

0

-10

-10

-20

-20

-30

-30

-40

-40 -50

-50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Patient number

E

10

0

0

-10

-10

-20

-20

-30

-30

-40

-40

-50

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

-50

volume reduction (cc)

volume and CTN reduction of NST 10

1

1.2

50 40 30 contralateral parotid ipsilateral parotid Linear fit (contralateral parotid) Linear fit (ipsilateral parotid)

20 10 0 0

10

0.4 0.6 0.8 volume reduction ratio (%)

-10

volume and CTN reduction of contralateral parotid gland

-50

Mean CTN reduction (Hu)

10

volume reduction (cc)

CTN reduction (Hu)

volume and CTN reduction of ipsilateral parotid gland

Patient number

CTN reduction (Hu)

GTV Linear fit (GTV)

B

B

CTN reduction (Hu)

35 30 25 20 15 10 5 0 -5 -10 0

mean CTN reduction mean mode reduction volume reduction

Patient number

Fig. 1. The volume, mean, and mode CT number (CTN) reduction of (a) gross tumor volume (GTV), (b) ipsilateral parotid gland, (c) contralateral parotid gland, (d) spinal cord, and (e) nonspecified tissue (NST) between the first and last radiation fraction. Figure 3. The GTV CTN histograms at fractions 1, 7, 13, 21, 28, and 35 are compared in Figure 3a, where the inset shows a zoomed view for fractions 21, 28, and 35. The absolute and relative mean CTN changes in GTV, contraand ipsilateral parotid glands, spinal cord, and NST as a function of mean dose delivered to each structure during the course of RT are displayed in Figure 3b and c,

0.2 0.4 0.6 volume change ratio (%)

0.8

Fig. 2. Pearson analysis between the volume reduction ratio and mean CT number (CTN) reduction for (a) gross tumor volume (GTV) and (b) parotid glands among all the patients. respectively. Similar data for one of the patients (patient 2) who had substantial CTN changes in parotid glands are presented in Figure 4. It is seen from Figures 3a and 4a that the peaks of CTN histograms in the GTV gradually shifted to the lower CTNs with increase of delivered doses (treatment fractions). The absolute and relevant mean CTNs in GTV and parotid glands were reduced with increase of delivered doses for these 2 patients (Figs. 3b, c and 4b, c). For patient 10, the absolute mean CTN changes in GTV and contralateral parotid gland were strongly correlated with the delivered dose, whereas the correlation was weak for the ipsilateral parotid gland (Fig. 3b, c). The corresponding correlation coefficients and P values are rZ0.98 and P<.001, rZ0.68 and PZ.109, and rZ0.98 and P<.001, respectively, for GTV and ipsi- and contralateral parotid glands. For patient 2 (Fig. 4), the absolute mean CTN changes for both ipsi- and contralateral parotid glands had a fair correlation with the increased doses, whereas the correlation for GTV was weak. The correlation coefficients and P values were rZ0.56 and PZ.226, rZ0.91 and PZ.005, and rZ0.83 and PZ.027, respectively, for GTV and ipsi- and contralateral parotid glands. Figure 5 shows the CTN changes as a function of delivered fractions in 5 regions within a parotid gland receiving doses of <5, 5 to 20, 20 to 30, 30 to 40, and 40 to 50 Gy after delivery of the whole RT course for patient 7.

Discussion The significant anatomic changes (eg volume regressions of the tumors and parotid glands) observed in this study during RT for HNC are generally comparable to those reported previously (1, 2, 9-12). For example, it has been reported

Volume 91  Number 5  2015

CT number change 1045

A

A 1500

50 0 0

2500

Fraction 1 Fraction 7 Fraction 13 Fraction 21 Fraction 28 Fraction 35

100

Fraction 1 Fraction 7 Fraction 13 Fraction 21 Fraction 28 Fraction 35

2000

1000

Count

Count

1500

1000

500 500

0

0 -200

-40

-20

0

20

40

60

80

40 20 GTV Contrallateral parotid Ipsilateral parotid Spinal cord NST linear fit (GTV)

0 -20 -40 -60

absolute mean CTN (Hu)

absolute mean CTN (Hu)

60

50

100

Contralateral parotid Ipsilateral parotid Spinal cord NST linear fit(Ipsilateral parotid)

2000

4000

6000

8000

Radiation dose (cGy)

8000

C 10

20

0

0 -20 GTV Contralateral parotid Ipsilateral parotid Spinal cord linear fit (GTV)

-40 -60 -80

relevant mean CTN (Hu)

relevant mean CTN (Hu)

C

0

GTV

0

2000 4000 6000 Radiation dose (cGy)

-50 CTN

50 40 30 20 10 0 -10 -20 -30 -40 -50

-80 0

-100

B

CTN

B

-150

100

-10 -20 GTV

-30

Contralateral parotid

-40

Ipsilateral parotid

-50

Spinal cord

-60

linear fit (Ipsilateral parotid)

-70 -80

0

-100

2000

4000

6000

8000

Radiation dose (cGy)

0

2000 4000 6000 Radiation dose (cGy)

8000

Fig. 3. Details of CT number (CTN) changes during the entire course of radiation for patient 10. (a) The CTN histograms of gross tumor volume (GTV) at fractions 1, 7, 13, 21, 28, and 35, from upper to lower dashed lines. The Inset shows the zoom-in figures for the fractions 21, 28 and 35. (b) Absolute mean CTN changes in GTV, contra- and ipsilateral parotid glands, spinal cord, and nonspecified tissue (NST) according to the delivered dose. (c) Relative mean CTN changes in GTV, contra- and ipsilateral parotid glands, and spinal cord NST according to the delivered dose. that the average volume reduction was 43.5% and 44.0% for the ipsilateral and contralateral parotid glands during IMRT for HNC patients (9). The CTN changes observed in both GTV and normal parotid glands during the course of RT delivery are highly patient specific: the substantial changes were not observed in all patients, although they received similar radiation doses, implying that the radiation

Fig. 4. Details of CT number (CTN) changes during the entire course of radiation for patient 2. (a) The CTN histograms of gross tumor volume (GTV) at fractions 1, 7, 13, 21, 28, and 35 from upper to lower dashed lines. (b) Absolute mean CTN changes in GTV, contra- and ipsilateral parotid glands, spinal cord, and nonspecified tissue (NST) according to the delivered dose. (c) Relative mean CTN changes in GTV, contra- and ipsilateral parotid glands, and spinal cord NST according to the delivered dose. dose may be just one of many factors causing the CTN changes. For the cases with substantial CTN changes, the mean CTN changes in the GTV or OARs were correlated with the radiation doses received. Changes to CTN in tumors and normal structures during and after RT with either SBRT (5-7) or conventional fractionations (8, 13-19) have been reported in several tumor sites, mostly for lung tumors. For example, Diot et al (5) observed that CTN of normal lung tissue increased with

International Journal of Radiation Oncology  Biology  Physics

1046 Feng et al.

A

Fraction numbers -20

Mean CTN (Hu)

-25 0

10

20

30

-30 -35

40 Right parotid (high dose) Left parotid (low dose)

-40 -45 -50 -55 -60

B

Left parotid gland

Fraction numbers -25

Mean CTN (Hu)

-30 0 -35 -40 -45 -50

10

20

30

40 5 Gy 20 Gy 30 Gy 40 Gy 50 Gy

-55 -60

Fig. 5. The CT number (CTN) changes as a function of delivered fractions in (a) the both sides of parotid glands and (b) 5 regions within a parotid gland receiving doses of <5, 5-20, 20-30, 30-40, and 40-50 Gy after delivery of the whole radiation therapy course for patient 7. radiation dose until 35 Gy and became unchanged thereafter during SBRT for lung tumors. With 3, 18, 24, and 30 months’ follow-up after RT, the rate of CTN increase with dose remained approximately 0.24% per gray (5). Mayer et al (8) observed CTN reductions of 3 to 36 HU in lung tumors treated with conventionally fractionated RT of 55.8 to 66.6 Gy. They reported a case in which the RT with 39.6 Gy in 22 fractions resulted in reductions of tumor volume from 70 cm3 to 30 cm3 and the mean CTN from 28 HU to 15 HU, whereas no correlation between the 2 reductions was observed (8). Recently, Xu et al (13) observed the CTN changes in nasopharyngeal cancer patients during the course of RT with tomotherapy based on megavoltage (MV) CTs. De et al (17) reported that the CTN change with radiation dose in lung tumors was clearly patient specific and ranged from 0 to 10 HU/Gy. In liver irradiation, Howell et al (7) reported that the reduced post-SBRT normal liver CTN was correlated with the radiation dose received. Thalacker et al (19) observed that the CTN of white matter was reduced by 5 HU after brain irradiation. Compared with most of these previous studies (5-8, 14-19) reporting CTN changes after RT in selected normal structures according to follow-up CTs after RT, the present study discovers CTN change during the course of RT delivery for HNC for both GTV and OARs. Although CTN changes during and after RT could be measured using various forms of CT (eg, kilovoltage or MV cone-beam CT, MV CT [Tomotherapy CT]), CTN measurement with diagnosticquality CT (eg, CT-on-rails as used in this work) should be the most accurate. The mechanisms for radiation-induced CTN changes are still unclear. It has been reported that an increase in blood

volume was observed in the primary tumor volume early in the course of RT (after 2 weeks of RT) in the HNC patients and that the increase of blood volume for patients with local control (median change, 5.1 mL/100 g) was significantly higher than that for patients with local failure (median change of 1.0 mL/100 g) (20). This study indicated that increased blood volume might be an early factor to predict the prognosis of HNC patients. According to that study, we inferred that increased tumor blood volume might contribute to the CTN reduction in GTV observed in this study. With increasing radiation dose, the tumor volume began to shrink, whereas the tumor blood volume might begin to increase. Increased tumor blood volume might cause the tumor density shift to hypodense, resulting in the CTN reduction in GTV. Another contributing factor may be that, with the increasing radiation dose, the tumor began to shrink as a result of some tumor cell death, which may eventually reduce CTN. More detailed studies are certainly needed to verify these hypotheses and to explore more substantial mechanisms. It is even more difficult to understand the radiationinduced CTN reduction in parotid glands, which may be contributed to by multiple factors. Histologically, the salivary gland consisted of acini, ducts, and adipose and fibrovascular tissue (21). Among them, the content of the adipose tissue showed individual variations in the same age group and an age-dependent increase (22, 23), and affected the size and the CTN of the salivary gland. Heo et al (21) reported that the CTN of the normal salivary gland increased with age. Teshima et al (24) considered there was a correlation between decreased parotid gland volume and decreased saliva production in patients with HNC undergoing RT. Different saliva production might be attributed to different saliva content in parotid glands. Stephens et al (25) studied morphologic changes in primates after irradiation with doses ranging from 2.5 to 15 Gy and reported that the number of degenerating and necrotic cells increased with dose. All these factors, including the decreased saliva production (24) and the increased number of degenerating and necrotic cells (25) with radiation dose, would contribute to the CTN change of parotid glands in HNC patients during and after RT. These multiple and complicated factors may contribute to the complex dose dependence observed in the CTN reduction of parotid glands. It is seen from Figure 5 that the CTNs of parotid glands first drop monotonically with lower doses, then increase with higher doses. When partitioning the parotid gland into subvolumes according to the doses received, the CTN of a high-dose subvolume drops at early fractions and increases in the middle of treatment, whereas the CTN of a low-dose subvolume decreases until the late part of treatment (Fig. 5). The CTN reduction varied from 7 HU to 25 HU for the case in Figure 5 and follows a complex manner with radiation dose. It has been reported that inflammatory salivary diseases were associated with higher-than-normal CTN (21), which may be one of the reasons for the increased CTN with higher radiation doses. However, the

Volume 91  Number 5  2015

exact mechanism behind the complex behavior between the CTN changes and radiation doses is still unclear. The radiation-induced CTN changes may be considered as an early indicator for radiation effect for a subset of patients, as evident in some tumor sites. Animal studies showed that CTN changes could directly relate to necrosis/ edema within brain tumors (26). The mean lung density (CTN) at the level of the left heart ventricle measured at 3 months after RT for breast cancers was increased significantly in patients with pneumonitis as compared with those without pneumonitis (67.0 vs 122.4 HU, PZ.001), whereas its value at 1 year after RT was increased in those patients with fibrotic lung changes as compared with those without (31.5 vs 91.8 HU, PZ.0001) (27). Another study showed that the change of 0.5 HU/Gy in normal lung tissue density was found to be correlated with lung toxicity (dyspnea grade 2) after RT for lung cancer (17). Chiou et al (28) found the sequential CTN changes might be correlated with the pathogenesis of veno-occlusive disease in hepatocellular carcinoma patients after hepatic irradiation. With this indirect evidence, it may be reasonable to assume the correlation of CTN change with some radiation effect in HNC patients after RT.

Conclusion The CT numbers in tumor and parotid glands can be reduced during the course of RT delivery for HNC. The reductions of mean CTN in GTV and parotid glands were weakly correlated with volume reductions, whereas they were strongly correlated with the radiation doses received for a subset of patients. The importance of these results for clinical practice is not clear. More detailed studies are needed to understand mechanism for the radiation-induced CTN changes, as well as their correlation with radiation effects.

CT number change 1047

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

References 21. 1. Barker JL Jr., Garden AS, Ang KK, et al. Quantification of volumetric and geometric changes occurring during fractionated radiotherapy for head and neck cancer using an integrated CT/linear accelerator system. Int J Radiat Oncol Biol Phys 2004;59:960-970. 2. Hansen EK, Bucci MK, Quivey JM, et al. Repeat CT imaging and replanning during the course of IMRT for head and neck cancer. Int J Radiat Oncol Biol Phys 2006;64:355-362. 3. Li XA. Adaptive Radiation Therapy. 1st ed. Boca Raton, FL: CRC Press; 2011. 4. Yang C, Liu F, Ahunbay E, et al. Combined online and offline adaptive radiation therapy: A dosimetric feasibility study. Pract Radiat Oncol 2014;4:e75-e83. 5. Diot Q, Kavanagh B, Schefter T, et al. Regional normal lung tissue density changes in patients treated with stereotactic body radiation therapy for lung tumors. Int J Radiat Oncol Biol Phys 2012;84:10241030. 6. Palma DA, van So¨rnsen de Koste J, Verbakel WF, et al. Lung density changes after stereotactic radiotherapy: A quantitative analysis in 50 patients. Int J Radiat Oncol Biol Phys 2011;81:974-978. 7. Howells CC, Stinauer MA, Diot Q, et al. Normal liver tissue density dose response in patients treated with stereotactic body radiation

22.

23. 24.

25.

26.

27.

28.

therapy for liver metastases. Int J Radiat Oncol Biol Phys 2012;84: e441-e446. Mayer R, Stanton K, Kleinberg L, et al. CT number distribution and its association with local control and as a marker of lung tumor response to radiation. Radiat Oncol Investig 1998;6:281-288. Fiorentino A, Caivano R, Metallo V, et al. Parotid gland volumetric changes during intensity-modulated radiotherapy in head and neck cancer. Br J Radiol 2012;85:1415-1419. Nishimura Y, Nakamatsu K, Shibata T, et al. Importance of the initial volume of parotid glands in xerostomia for patients with head and neck cancers treated with IMRT. Jpn J Clin Oncol 2005;35:375-379. Loo H, Fairfoul J, Chakrabarti A, et al. Tumour shrinkage and contour change during radiotherapy increase the dose to organs at risk but not the target volumes for head and neck cancer patients treated on the TomoTherapyHiArt system. Clin Oncol (R Coll Radiol) 2011;23:40-47. Va´squez Osorio EM, Hoogeman MS, Al-Mamgani A, et al. Local anatomic changes in parotid and submandibular glands during radiotherapy for oropharynx cancer and correlation with dose, studied in detail with nonrigid registration. Int J Radiat Oncol Biol Phys 2008; 70:875-882. Xu SP, Wu ZX, Yang CG, et al. Correlation of changes in CT number and volume of gross tumor volume during the course of radiation therapy for nasopharyngeal cancer. Abstract presented at Annual Meeting of Chinese Medical Physics Association, May 15-17, 2014, Xi’an, China. Ma J, Zhang J, Zhou S, et al. Regional lung density changes after radiation therapy for tumors in and around thorax. Int J Radiat Oncol Biol Phys 2010;76:116-122. Bernchou U, Schytte T, Bertelsen A, et al. Time evolution of regional CT density changes in normal lung after IMRT for NSCLC. Radiother Oncol 2013;109:89-94. Dale E, Ha˚rsaker V, Kristoffersen DT, et al. CT density in lung cancer patients after radiotherapy sensitized by metoclopramide. A subgroup analysis of a randomized trial. Strahlenther Onkol 2010;186:163-168. De RD, Sharifi H, Defraene G, et al. Quantification of radiationinduced lung damage with CT scans: The possible benefit for radiogenomics. Acta Oncol 2013;52:1405-1410. Phernambucq EC, Palma DA, Vincent A, et al. Time and dose-related changes in radiological lung density after concurrent chemoradiotherapy for lung cancer. Lung Cancer 2011;74:451-456. Thalacker U, Liszka G, Somogyi A, et al. The observation of edema in the substantia alba during postoperative brain irradiation. The role of computed tomographic studies. Strahlenther Onkol 1998;174:14-18. Yue C, Aron P, Diana L, et al. Early prediction of outcome in advanced head-and-neck cancer based on tumor blood volume alterations during therapy: A prospective study. Int J Radiat Oncol Biol Phys 2008;72: 1287-1290. Heo MS, Lee SC, Lee SS, et al. Quantitative analysis of normal major salivary glands using computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001;92:240-244. Scott J, Flower EA, Burns J. A quantitative study of histological changes in the human parotid gland occurring with adult age. J Oral Pathol 1987;16:505-510. Liu X, Yu G, Tang G. Age changes in CT values of human major salivary glands. Zhong Hua Kou Qiang Yi Xue Za Zhi 1999;34:73-75. Teshima K, Murakami R, Tomitaka E, et al. Radiation-induced parotid gland changes in oral cancer patients: Correlation between parotid volume and saliva production. Jpn J Clin Oncol 2010;40:42-46. Stephens LC, Ang KK, Schultheiss TE, et al. Target cell and mode of radiation injury in rhesus salivary glands. Radiother Oncol 1986;7: 165-174. Fike JR, LeCouteur RA, Cann CE, et al. Computerized tomography of brain tumors of the rostral and middle fossas in the dog. Am J Vet Res 1981;42:275-281. Kaha´n Z, Csenki M, Varga Z, et al. The risk of early and late lung sequelae after conformal radiotherapy in breast cancer patients. Int J Radiat Oncol Biol Phys 2007;68:673-681. Chiou SY, Lee RC, Chi KH, et al. The triple-phase CT image appearance of post-irradiated livers. Acta Radiol 2001;42:526-531.