Abstract
Background/Aim: We evaluated the influence of previous treatments on the parametric discrepancies between dose–volume histograms (DVHs) and dose–function histograms (DFHs) generated based on 99mTc-GSA SPECT images of hepatocellular carcinoma (HCC) patients treated with stereotactic body radiation therapy (SBRT). Patients and Methods: Twelve patients underwent SBRT at 30-40 Gy. Registration between planning CT and SPECT/CT images was performed, and DFH parameters were calculated as follows: Fx=(sum of the counts within the liver volume receiving a dose of more than x Gy/sum of the counts within the whole liver volume) ×100. The discrepancy between Fx and Vx (Dx) was also calculated. Results: The number of previous treatments for lesions other than SBRT-treated lesions (≥2 vs. <2) exhibited a significant influence on the absolute values of D10, D15, and D20 (p<0.05). Conclusion: Previous treatment significantly influences the parametric discrepancy between DFH and DVH.
- Hepatocellular carcinoma
- stereotactic body radiation therapy
- dose–function histogram
- dose–volume histogram
- radiation-induced liver disease
Primary liver cancer is the third most common cause of cancer death worldwide. Hepatocellular carcinoma (HCC) accounts for 85%-90% of primary liver cancers (1). Stereotactic body radiation therapy (SBRT) for HCC has been introduced as an alternative to standard treatments such as surgical resection and radiofrequency ablation (RFA) (1). SBRT delivers a highly conformal, potent dose of radiation to the tumor in some fractions, while minimizing radiation damage to organs at risk. SBRT provides excellent local control for HCC with a reported control rate of 80-90% (2-4).
HCC response to radiation therapy (RT) exhibits a dose–response relationship (5), but a fine balance is required between delivering a sufficient RT dose to control the HCC and avoiding radiation-induced liver toxicity. Radiation-induced liver injury (RILD), which occurs in 10-20% of HCC patients undergoing SBRT, remains a problematic adverse effect, because of pre-existing liver dysfunctions occurring secondary to comorbid conditions such as hepatitis B/C infection and cirrhosis (3, 4). The percentage of normal liver volume receiving a dose over the threshold dose calculated based on a dose–volume histogram (DVH) is commonly used for determining dose constraints and predicting radiation-induced liver toxicity; however, it has limited predictive ability (6, 7).
A drawback of using DVH for SBRT planning in HCC is the lack of functional information because DVHs are generated based on computed tomography (CT) images, which provide only morphological information. A dose–function histogram (DFH) generated based on single photon emission computed tomography (SPECT) images may facilitate the assessment of the functional status of organs at risk. For treatment planning in lung cancers, Marks et al. (8) performed DFH calculation using 99mTc-labeled macro-aggregated albumin and observed decreases in lung perfusion at, adjacent to, and separate from tumor in 94%, 74%, and 42% of patients, respectively. 99mTc-labeled diethylene triamine pentaacetate-galactosyl human serum albumin (99mTc-GSA) SPECT provides three-dimensional information of regional liver function (9). 99mTc-GSA SPECT findings suggest that regional function of patients with liver tumors is inhomogeneous due to previous treatments such as RFA and transarterial chemoembolization (TACE) (10), causing discrepancy between DFH and DVH parameters. However, to our knowledge, DFH calculation using fused images from SPECT and planning CT for SBRT for HCC patients has not been reported.
Patient characteristics.
This study evaluated discrepancy between DFH and DVH parameters and influence of previous treatments on parametric discrepancy for HCC patients who underwent SBRT.
Patients and Methods
Patients. This retrospective study based on prospectively acquired data received institutional review board approval. Written informed consent was obtained from all patients. Between August 2013 and February 2017, 12 consecutive HCC patients (nine men and three women; mean age, 68.8 years; range=58-79 years) were included in this study (Table I). Surgery had been performed in three (25%) patients before treatment. Nine (75%) patients had undergone RFA, TACE, and/or percutaneous ethanol injection therapy (PEIT) for HCC in the liver or remnant liver, and no patient received SBRT before treatment. The mean number of previous treatments for lesions other than SBRT-treated lesions was 1.8 (range=0-4 treatments). All patients underwent 99mTc-GSA SPECT/CT imaging within 1 month before SBRT planning.
SPECT/CT imaging. We used a SPECT/CT system with dual-head detectors and a 16-row multidetector CT scanner (Symbia T16, Siemens Healthcare, Erlangen, Germany). Hepatic SPECT data (60 steps of 15 s/step, 360°, 128×128 matrix) were obtained using a gamma camera fitted with a low-to-medium energy general-purpose collimator at 20-35 min after the intravenous injection of 99mTc-GSA (185 MBq). We acquired 64 projections at 6° intervals in the continuous mode. To enable SPECT attenuation correction, non-contrast-enhanced helical CT images were obtained. The scanning parameters were as follows: tube voltage, 120 kVp; tube current, 200-275 mA; rotation time, 0.7 sec; beam collimation, 1.25 mm; beam pitch, 0.87; field of view, 36 cm; matrix, 512×512 pixels; slice thickness, 2 mm; and slice interval, 2 mm. After registration between SPECT and CT images, a CT-derived attenuation-coefficient map was created. For SPECT reconstruction, the ordered-subset expectation maximization algorithm (Flash 3D, Siemens Healthcare, Erlangen, Germany) was used (eight iterations; six subsets). Postprocessing was performed with a 7.8-mm Gaussian filter.
Planning simulation CT and treatment planning. The details of the planning simulation CT and treatment planning are described elsewhere (11). Briefly, we used a LightSpeed RT CT scanner (GE Medical Systems, Waukesha, WI, USA). Patients lay supine, and abdominal compression was applied. Immediately after a dynamic contrast-enhanced scan, a non-helical slow-speed scan with a gantry rotation time of 4 s, slice thickness of 2.5 mm, and slice interval of 2.5 mm was performed.
For planning simulation, the images from the dynamic contrast-enhanced scan and slow-speed scan were exported to the Pinnacle3 treatment planning system (version 9.2; Phillips Radiation Oncology Systems, Fitchburg, WI, USA) and were registered by hardware arrangement. The structures of the target and organs at risk were delineated based on the slow-speed CT images. Treatment planning was performed using 8-9 coplanar and non-coplanar fields with a 6-MV and/or 10-MV photon beam. A total RT dose of 30-40 Gy in five fractions was prescribed at the 70%-80% isodose line of the maximum dose (Table I). The percentage of normal liver volume (liver – gross tumor volume) receiving a dose of more than 20 Gy (V20) did not exceed 20%. SPECT images were not used for treatment planning.
Calculating the parameters of DFH and DVH. Attenuation-corrected SPECT and CT images and planning CT images, including delineated structures and dose distributions, were transferred to Velocity AI (version 3.0.2; Varian Medical Systems, Palo Alto, CA, USA). After registration between SPECT and CT images by hardware arrangement, we registered SPECT/CT images onto the planning CT images: a rigid image registration followed by a non-rigid deformable registration. During this process, each SPECT voxel was mapped to a new position based on the transformations used in the CT–CT registration, resulting in new SPECT/CT fused images that were deformably registered with the planning CT images (12). Structures of the irradiated volumes of the liver parenchyma were generated at 5-Gy dose increments, based on the dose distribution information.
Difference between Fx and Vx (Dx).
DFH parameters were calculated as follows:
DVH parameters were calculated as follows:
Statistical analysis. The difference between Fx and Vx (Dx) was calculated by subtracting Vx from Fx (i.e., Fx – Vx) for each parameter. The effect of the number of previous treatments (of RFA, TACE, and PEIT) for lesions other than SBRT-treated lesions on the absolute value of Dx was evaluated using the Mann–Whitney U-test. Differences with p-values of <0.05 were considered statistically significant. Statistical calculations were performed with SPSS software, version 24.0 (SPSS Inc., Chicago, IL, USA).
Results
Dx% ranged from −3.4% to 6.8% (Table II). Dx was positive for all parameters for five (42%) patients and ranged from negative to positive for seven (58%) patients. The mean absolute values of Dx were 2.66 and 1.69 at D5 (p=0.755), 3.22 and 1.67 at D10 (p=0.048), 2.86 and 1.11 at D15 (p=0.018), 2.00 and 0.77 at D20 (p=0.048), 1.32 and 0.71 at D25 (p=0.106), 0.96 and 0.75 at D30 (p=0.343), and 0.74 and 0.72 at D35 (p=1.000) for ≥2 and <2 previous treatments, respectively, of lesions other than SBRT-treated lesions (Figure 1). Figures 2 and 3 show representative cases with large parametric discrepancies between DFH and DVH parameters.
Discussion
Because HCC patients often suffer from metachronous multicentric occurrence or intrahepatic metastasis and RT is not considered the first choice of treatment for HCC, most patients undergo other treatments, such as RFA and TACE, before RT (5, 13). A history of prior treatment may cause inhomogeneity in liver function (10). Our study revealed a discrepancy between DFH and DVH parameters and suggested that previous treatments influenced this parametric discrepancy.
The geometric relationship between the irradiated liver and previously treated lesions may influence the magnitude relationship between DFH and DVH parameters. Theoretically, when previous treatments are performed for lesions distant from SBRT-treated lesions and radioisotope (RI) non-accumulation occurs outside the irradiated liver parenchyma, the value of DFH parameters are greater than those of corresponding DVH parameters. In such cases, the risk of radiation-induced liver toxicity should be underestimated with DVH. By contrast, when previous treatments are performed for lesions close to SBRT-treated lesions and RI non-accumulation occurs within the irradiated liver parenchyma, the value of some DFH parameters is less than that of corresponding DVH parameters. In such cases, the risk of radiation-induced liver toxicity should be overestimated with DVH. In this study, the magnitude relationship between DFH and DVH parameters varied according to the doses, even for the same patients. In clinical settings, patients may receive multiple treatments for multiple sites of disease, and a lack of RI accumulation may be observed at various sites in the liver parenchyma. Hence, we could not identify a clear trend for the magnitude relationship between DFH and DVH parameters.
Mean absolute values of the difference between Fx and Vx (Dx) according to the number of previous treatments of radiofrequency ablation, transarterial chemoembolization, and percutaneous ethanol injection therapy for lesions other than stereotactic body radiation therapy-treated lesions.
Dosimetric factors associated with toxicity are not well established for HCC patients treated with hypofractionated RT. Son et al. (14) treated 72 HCC patients using helical tomotherapy at 40–50 Gy in 10 fractions. They found that V15 was significantly associated with an increase in the Child–Pugh score. Liang et al. (15) treated 114 primary liver carcinomas using hypofractionated conformal RT at 40-68 Gy with a fraction size of 4-6 Gy. They found that V20 was an independent predictor of RILD. Dyk et al. (16) treated 23 patients with primary liver carcinomas using SBRT at 36-60 Gy in 3-6 fractions. They found that V25 was a predictor of decline in liver function after treatment. Although these reports evaluated patients with similar RT doses and fraction sizes, the recommended DVH parameters varied. The RT dose and fraction size of our patients were similar to those described in these previous reports, and the recommended parameters overlapped with the parameters of the significant discrepancy between DFH and DVH observed in our study. One possible reason for the variability in the recommended DVH parameters of the previous reports may be that the inhomogeneity of liver function was not considered. The use of DFH may facilitate the detection of dosimetric parameters for the precise prediction of radiation-induced liver toxicity.
Our study has certain limitations, including the relatively small number of patients who were examined. The previous treatments included RFA, TACE, and PEIT. We could not evaluate the differences in the influence of each treatment modality on the parametric discrepancy between DFH and DVH. We compared DFH and DVH parameters from a dosimetric perspective but could not compare the predictability of these parameters for radiation-induced liver toxicity. Further investigations are underway to address these issues.
In conclusion, previous treatments significantly influenced the parametric discrepancy between DFH and DVH. The use of DFH may facilitate the detection of dosimetric parameters for the precise prediction of radiation-induced liver toxicity.
A 63-year-old man with hepatocellular carcinoma (Patient 2). (A) Axial contrast-enhanced computed tomography (CT) image. He received radiofrequency ablation therapy twice for hepatocellular carcinoma in segment 8. (B) Fused image of single photon emission computed tomography (SPECT) and planning CT with the dose distribution of stereotactic body radiation therapy (SBRT). SPECT showed a decreased accumulation of the radioisotope around the treated area. We performed SBRT at 40 Gy for another lesion in segment 4. (C) The values of dose–function and dose–volume histogram parameters. F15 and V15 were 9.4% and 12.8%, respectively.
A 79-year-old man with hepatocellular carcinoma (Patient 8). (A) Axial contrast-enhanced computed tomography (CT) image. He received transarterial chemoembolization and radiofrequency ablation for hepatocellular carcinoma in segment 3 and hepatic resection of segment 6. (B) Fused image of single photon emission computed tomography (SPECT) and planning CT with the dose distribution of stereotactic body radiation therapy (SBRT). SPECT showed a decreased accumulation of the radioisotope around the treated area in segment 3. We performed SBRT at 35 Gy for a recurrent lesion near the resected area of the remnant liver. (C) The values of dose–function and dose–volume histogram parameters. F10 and V10 were 42.7% and 36.0%, respectively.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number 26861004.
The Authors wish to thank Ryuji Ikeda and Masato Maruyama, Kumamoto University Hospital, for their technical supports.
Footnotes
This article is freely accessible online.
Conflicts of Interest
None.
- Received December 26, 2017.
- Revision received January 22, 2018.
- Accepted January 24, 2018.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
