Original Study
FDG-PET Predicts Pain Response and Local Control in Palliative Radiotherapy With or Without Systemic Treatment in Patients With Bone Metastasis From Nonesmall-cell Lung Cancer Fen Zhao,1,2 Gang Ding,3 Wei Huang,1,2 Minghuan Li,1,2 Zheng Fu,4 Guoren Yang,4 Li Kong,1,2 Yan Zhang,1,2 Jinming Yu1,2 Abstract The purpose of the present study was to evaluate the prognostic value of the maximal standardized uptake value (SUVmax) from serial positron emission tomography scans in patients with bone metastases from nonesmall-cell lung cancer. The results showed that the pre-RT SUVmax predicted the initial pain severity and local control. Moreover, the change in the SUVmax after palliative radiotherapy predicted the pain response and local control rate. Introduction: We sought to evaluate the value of fluorine-18 fluorodeoxyglucose positron emission tomography (PET) in predicting the pain severity, pain response, and in-field tumor control after palliative radiotherapy (RT) in patients with nonesmall-cell lung cancer (NSCLC) bone metastases. Materials and Methods: The present retrospective, institutional review boardeapproved study involved 74 patients with NSCLC and 185 bone metastatic lesions. All patients had undergone PET-computed tomography (CT) scans before and after RT. The pain scores were determined using a numerical rating scale, and the maximal standardized uptake value (SUVmax) at each location was recorded. The pain scores and responses to RT were compared using the pre-RT SUVmax and SUVmax changes after RT. Cox regression analyses were performed to identify the prognostic factors for in-field progression-free survival (PFS) and in-field event-free survival (EFS). Results: The pre-RT SUVmax correlated with the initial pain scores (r ¼ 0.885, P < .001), and the decrease in the SUVmax after RT was associated with the pain response to RT (P ¼ .001). During the follow-up period, 47.03% and 38.92% of the lesions showed in-field tumor radiographic progression and in-field events, respectively. The Cox regression analyses showed that a higher pre-RT SUVmax ( 8.2) was an independent prognostic factor of worse in-field PFS and worse in-field EFS (hazard ratio [HR] 1.42 and 1.46; P ¼ .044 and P ¼ .005, respectively) and that a greater SUVmax decrease ( 28.3%) after RT was an independent prognostic factor of better in-field PFS and better in-field EFS (HR 0.59 and 0.60, respectively; P < .001 for both). Conclusion: In patients with NSCLC osseous metastasis treated with palliative RT, the pre-RT SUVmax predicted the initial pain severity and local control. Moreover, the change in the SUVmax after RT predicted the pain response and local control. Clinical Lung Cancer, Vol. -, No. -, --- ª 2015 Elsevier Inc. All rights reserved. Keywords: Fluorine-18 fluorodeoxyglucose, Positron emission tomography, RT, Prognosis, Standard uptake value
1 Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Shandong University, Jinan, Shandong, China 2 Key Laboratory of Radiation Oncology of Shandong Province, Shandong Cancer Hospital and Institute, Jinan, Shandong, China 3 Department of Ophthalmology, Jinan Second People’s Hospital, Jinan, Shandong, China 4 Department of Nuclear Medicine, Shandong Cancer Hospital and Institute, Jinan, Shandong, China
1525-7304/$ - see frontmatter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cllc.2015.01.005
Submitted: Aug 19, 2014; Revised: Jan 18, 2015; Accepted: Jan 23, 2015 Address for correspondence: Jinming Yu, MD, PhD, Department of Radiation Oncology, Shandong Cancer Hospital and Institute, 440 Jiyan Road, Jinan, Shandong 250117, China E-mail contact: [email protected]
Clinical Lung Cancer Month 2015
-1
Prognostic Value of
18
F-FDG PET in NSCLC Bone Metastasis
Introduction
2
-
metastases; (6) a pain score < 2 on a scale of 0 to 10; (7) bone metastases identified using bone scintigraphy or magnetic resonance imaging, with no corresponding FDG uptake evident; or (8) the follow-up data after RT were incomplete (no records of status after RT completion, including interval to radiographic progression or interval to an in-field event). Thus, 74 patients with NSCLC with 185 (96 spine and 89 nonspine metastases) measurable osteolytic bone metastatic lesions were included in the present study. All metastatic lesions and other clinicopathologic characteristics are listed in Table 1. The Shandong Cancer Hospital and Institute Institutional Review Board approved our retrospective observational study. All patients provided informed consent for retaining their data for subsequent research.
Approximately 30% to 40% of patients with nonesmall-cell lung cancer (NSCLC) will develop bone metastases during the clinical disease course.1 Bone lesions caused by lung cancer metastasis are often osteolytic and mixed-type lesions and have been associated with a poorer prognosis than sclerotic metastasis.2 The ensuing complications caused by bone metastasis, referred to as skeletalrelated events (SREs), include intractable pain, spinal cord compression, pathologic fractures, hypocalcemia, and the need for radiation or surgery directed at the bone.3 These complications have been associated with significant deterioration in patients’ quality of life and can even be life-threatening. Local palliative radiotherapy (RT) is one of the most important treatment modalities for bone metastases.4 Approximately 50% to 80% of the patients with painful bone metastasis who undergo RT will experience partial relief and 20% to 50%, complete relief.5-7 However, the response of bone pain to palliative RT and the long-term in-field tumor control in patients with bone metastasis has varied among patients receiving the same treatment. An appropriate assessment of the treatment responses and early identification of treatment failure are essential for making the correct treatment decisions and improving outcomes. Positron emission tomography (PET) with fluorine-18 (18F) fluorodeoxyglucose (FDG) integrated with computed tomography (CT) can assess tumor metabolic function and anatomic localization. Moreover, it has been shown to be more accurate than conventional imaging for the diagnosis of bone metastases in patients with lung cancer.8,9 FDG uptake is proportional to the growth rate and proliferation capacity of tumors and might therefore help in predicting the biologic aggressiveness of a tumor.10 The standardized uptake value (SUV), a parameter used to assess a tumor’s uptake of FDG on PET, has also been identified as an independent prognostic factor correlating with tumor aggressiveness, treatment outcomes, and survival in patients with lung cancer.11-15 However, despite the increasing use of integrated PET-CT in the management of lung cancer, to our knowledge, little has been reported about its role in the evaluation of treatment responses in terms of pain or its prognostic value for local control in patients with NSCLC and bone metastasis undergoing palliative RT. Thus, the purposes of our study were to evaluate whether FDG-PET avidity before or after RT can predict the pain responses and whether PET can be used to predict in-field local control.
The images were reviewed by 2 experienced nuclear medicine radiologists using an Xeleris workstation (GE Healthcare). The slice with the highest SUV and the 2 adjacent slices were typically included in the region of interest analysis to obtain the maximal SUV (SUVmax) per lesion. The change in SUVmax (DSUVmax) after treatment was calculated using the following equation: DSUVmax ¼ [(SUVpre SUVpost)/SUVpre] 100, where SUVpre and SUVpost indicate the pre- and post-treatment SUVmax, respectively. For the CT scans, the morphologic tumor sizes were measured unidimensionally in the longest transaxial section by 2 physicians.
Materials and Methods
RT Protocol
Patients
RT was performed using a linear accelerator. For spinal, pelvic, and long bone lesions, 6 or 15 MV of photon energy was used. For lesions in the ribs or sternum, 8 to 18 MeV electron energies were delivered. Two-dimensional (2D) RT was used for 101 lesions (64 total fields), and 3-dimensional (3D) conformal RT was used for 84 lesions (54 fields). Various fractionation schedules were used (3 Gy 10 fractions [F], 2.5 Gy 14 F, 2 Gy 20 F, or 8.0 Gy 1 F), and the relative frequencies with which they were used are listed in Table 1. The total doses and respective fractions were planned separately for each individual patient, depending on the patient’s general state of health, location of the metastatic lesion, prognosis, tumor histologic finding, and current staging. For the dose comparisons, the doses were converted to the equivalent dose
A total of 920 patients with osseous metastases from NSCLC, who had undergone palliative RT at Shandong Cancer Hospital and Institute from June 2006 to December 2012, were identified from the patient database. Cross-referencing the PET-CT database revealed that 184 of these patients had undergone FDG-PET. Of the 184 patients, 110 were excluded for the following reasons: (1) radiographic evidence of osteoblastic lesions (FDG-PET has a lower sensitivity for detecting osteoblastic lesions) before RT; (2) lytic lesions < 10 mm; (3) receipt of granulocyte colonystimulating factor therapy; (4) chemotherapy within 1 month of RT or previous RT at the same location; (5) pathologic fractures or spinal cord or cauda equina compression resulting from the bone
Clinical Lung Cancer Month 2015
PET-CT Scans PET-CT scans were performed before (mean, 6.6 2.8 days) and after (mean, 30.4 8.3 days) RT for each patient using an integrated PET-CT device (Discovery LS, GE Healthcare). The patients were instructed to fast for 4 to 6 hours before the injection of 18F-FDG. All patients were tested to confirm that their glucose level was within the normal range (80-120 mg/dL, 4.4-6.6 mmol/L) before FDG administration. After intravenous injection of 374.2 36.7 MBq 18 F-FDG over 40 to 60 minutes, a CT scan (parameters: 80 mA, 140 kV, 0.8 s/tube rotation, slice thickness 5 mm, data acquisition time 22.5 seconds) was performed from the head to the mid-thigh. A PET scan was acquired immediately thereafter using an acquisition time of 3 minutes per bed position. The CT data were used for attenuation correction, and the images were reconstructed using the ordered subsets expectation maximization algorithm.
Image Analysis
Fen Zhao et al Table 1 Patient (n [ 74) and Metastatic Lesion (n [ 185) Characteristics Characteristic
Number of Patients (% or Range)
Gender Male
46 (62.16)
Female
28 (37.84)
Median age (years)
57 (38-78)
Median no. of metastatic lesions
2 (1-5)
Median no. of treatment fields
1 (1-3)
Histologic type AC
38 (51.35)
SCC
22 (29.73)
ASC
7 (9.46)
LC
7 (9.46)
Karnofsky performance score
FDG-PET Predicts Pain Response and Local Control in Palliative Radiotherapy With or Without Systemic Treatment in Patients With Bone Metastasis From Nonesmall-cell Lung Cancer Fen Zhao,1,2 Gang Ding,3 Wei Huang,1,2 Minghuan Li,1,2 Zheng Fu,4 Guoren Yang,4 Li Kong,1,2 Yan Zhang,1,2 Jinming Yu1,2 Abstract The purpose of the present study was to evaluate the prognostic value of the maximal standardized uptake value (SUVmax) from serial positron emission tomography scans in patients with bone metastases from nonesmall-cell lung cancer. The results showed that the pre-RT SUVmax predicted the initial pain severity and local control. Moreover, the change in the SUVmax after palliative radiotherapy predicted the pain response and local control rate. Introduction: We sought to evaluate the value of fluorine-18 fluorodeoxyglucose positron emission tomography (PET) in predicting the pain severity, pain response, and in-field tumor control after palliative radiotherapy (RT) in patients with nonesmall-cell lung cancer (NSCLC) bone metastases. Materials and Methods: The present retrospective, institutional review boardeapproved study involved 74 patients with NSCLC and 185 bone metastatic lesions. All patients had undergone PET-computed tomography (CT) scans before and after RT. The pain scores were determined using a numerical rating scale, and the maximal standardized uptake value (SUVmax) at each location was recorded. The pain scores and responses to RT were compared using the pre-RT SUVmax and SUVmax changes after RT. Cox regression analyses were performed to identify the prognostic factors for in-field progression-free survival (PFS) and in-field event-free survival (EFS). Results: The pre-RT SUVmax correlated with the initial pain scores (r ¼ 0.885, P < .001), and the decrease in the SUVmax after RT was associated with the pain response to RT (P ¼ .001). During the follow-up period, 47.03% and 38.92% of the lesions showed in-field tumor radiographic progression and in-field events, respectively. The Cox regression analyses showed that a higher pre-RT SUVmax ( 8.2) was an independent prognostic factor of worse in-field PFS and worse in-field EFS (hazard ratio [HR] 1.42 and 1.46; P ¼ .044 and P ¼ .005, respectively) and that a greater SUVmax decrease ( 28.3%) after RT was an independent prognostic factor of better in-field PFS and better in-field EFS (HR 0.59 and 0.60, respectively; P < .001 for both). Conclusion: In patients with NSCLC osseous metastasis treated with palliative RT, the pre-RT SUVmax predicted the initial pain severity and local control. Moreover, the change in the SUVmax after RT predicted the pain response and local control. Clinical Lung Cancer, Vol. -, No. -, --- ª 2015 Elsevier Inc. All rights reserved. Keywords: Fluorine-18 fluorodeoxyglucose, Positron emission tomography, RT, Prognosis, Standard uptake value
1 Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Shandong University, Jinan, Shandong, China 2 Key Laboratory of Radiation Oncology of Shandong Province, Shandong Cancer Hospital and Institute, Jinan, Shandong, China 3 Department of Ophthalmology, Jinan Second People’s Hospital, Jinan, Shandong, China 4 Department of Nuclear Medicine, Shandong Cancer Hospital and Institute, Jinan, Shandong, China
1525-7304/$ - see frontmatter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cllc.2015.01.005
Submitted: Aug 19, 2014; Revised: Jan 18, 2015; Accepted: Jan 23, 2015 Address for correspondence: Jinming Yu, MD, PhD, Department of Radiation Oncology, Shandong Cancer Hospital and Institute, 440 Jiyan Road, Jinan, Shandong 250117, China E-mail contact: [email protected]
Clinical Lung Cancer Month 2015
-1
Prognostic Value of
18
F-FDG PET in NSCLC Bone Metastasis
Introduction
2
-
metastases; (6) a pain score < 2 on a scale of 0 to 10; (7) bone metastases identified using bone scintigraphy or magnetic resonance imaging, with no corresponding FDG uptake evident; or (8) the follow-up data after RT were incomplete (no records of status after RT completion, including interval to radiographic progression or interval to an in-field event). Thus, 74 patients with NSCLC with 185 (96 spine and 89 nonspine metastases) measurable osteolytic bone metastatic lesions were included in the present study. All metastatic lesions and other clinicopathologic characteristics are listed in Table 1. The Shandong Cancer Hospital and Institute Institutional Review Board approved our retrospective observational study. All patients provided informed consent for retaining their data for subsequent research.
Approximately 30% to 40% of patients with nonesmall-cell lung cancer (NSCLC) will develop bone metastases during the clinical disease course.1 Bone lesions caused by lung cancer metastasis are often osteolytic and mixed-type lesions and have been associated with a poorer prognosis than sclerotic metastasis.2 The ensuing complications caused by bone metastasis, referred to as skeletalrelated events (SREs), include intractable pain, spinal cord compression, pathologic fractures, hypocalcemia, and the need for radiation or surgery directed at the bone.3 These complications have been associated with significant deterioration in patients’ quality of life and can even be life-threatening. Local palliative radiotherapy (RT) is one of the most important treatment modalities for bone metastases.4 Approximately 50% to 80% of the patients with painful bone metastasis who undergo RT will experience partial relief and 20% to 50%, complete relief.5-7 However, the response of bone pain to palliative RT and the long-term in-field tumor control in patients with bone metastasis has varied among patients receiving the same treatment. An appropriate assessment of the treatment responses and early identification of treatment failure are essential for making the correct treatment decisions and improving outcomes. Positron emission tomography (PET) with fluorine-18 (18F) fluorodeoxyglucose (FDG) integrated with computed tomography (CT) can assess tumor metabolic function and anatomic localization. Moreover, it has been shown to be more accurate than conventional imaging for the diagnosis of bone metastases in patients with lung cancer.8,9 FDG uptake is proportional to the growth rate and proliferation capacity of tumors and might therefore help in predicting the biologic aggressiveness of a tumor.10 The standardized uptake value (SUV), a parameter used to assess a tumor’s uptake of FDG on PET, has also been identified as an independent prognostic factor correlating with tumor aggressiveness, treatment outcomes, and survival in patients with lung cancer.11-15 However, despite the increasing use of integrated PET-CT in the management of lung cancer, to our knowledge, little has been reported about its role in the evaluation of treatment responses in terms of pain or its prognostic value for local control in patients with NSCLC and bone metastasis undergoing palliative RT. Thus, the purposes of our study were to evaluate whether FDG-PET avidity before or after RT can predict the pain responses and whether PET can be used to predict in-field local control.
The images were reviewed by 2 experienced nuclear medicine radiologists using an Xeleris workstation (GE Healthcare). The slice with the highest SUV and the 2 adjacent slices were typically included in the region of interest analysis to obtain the maximal SUV (SUVmax) per lesion. The change in SUVmax (DSUVmax) after treatment was calculated using the following equation: DSUVmax ¼ [(SUVpre SUVpost)/SUVpre] 100, where SUVpre and SUVpost indicate the pre- and post-treatment SUVmax, respectively. For the CT scans, the morphologic tumor sizes were measured unidimensionally in the longest transaxial section by 2 physicians.
Materials and Methods
RT Protocol
Patients
RT was performed using a linear accelerator. For spinal, pelvic, and long bone lesions, 6 or 15 MV of photon energy was used. For lesions in the ribs or sternum, 8 to 18 MeV electron energies were delivered. Two-dimensional (2D) RT was used for 101 lesions (64 total fields), and 3-dimensional (3D) conformal RT was used for 84 lesions (54 fields). Various fractionation schedules were used (3 Gy 10 fractions [F], 2.5 Gy 14 F, 2 Gy 20 F, or 8.0 Gy 1 F), and the relative frequencies with which they were used are listed in Table 1. The total doses and respective fractions were planned separately for each individual patient, depending on the patient’s general state of health, location of the metastatic lesion, prognosis, tumor histologic finding, and current staging. For the dose comparisons, the doses were converted to the equivalent dose
A total of 920 patients with osseous metastases from NSCLC, who had undergone palliative RT at Shandong Cancer Hospital and Institute from June 2006 to December 2012, were identified from the patient database. Cross-referencing the PET-CT database revealed that 184 of these patients had undergone FDG-PET. Of the 184 patients, 110 were excluded for the following reasons: (1) radiographic evidence of osteoblastic lesions (FDG-PET has a lower sensitivity for detecting osteoblastic lesions) before RT; (2) lytic lesions < 10 mm; (3) receipt of granulocyte colonystimulating factor therapy; (4) chemotherapy within 1 month of RT or previous RT at the same location; (5) pathologic fractures or spinal cord or cauda equina compression resulting from the bone
Clinical Lung Cancer Month 2015
PET-CT Scans PET-CT scans were performed before (mean, 6.6 2.8 days) and after (mean, 30.4 8.3 days) RT for each patient using an integrated PET-CT device (Discovery LS, GE Healthcare). The patients were instructed to fast for 4 to 6 hours before the injection of 18F-FDG. All patients were tested to confirm that their glucose level was within the normal range (80-120 mg/dL, 4.4-6.6 mmol/L) before FDG administration. After intravenous injection of 374.2 36.7 MBq 18 F-FDG over 40 to 60 minutes, a CT scan (parameters: 80 mA, 140 kV, 0.8 s/tube rotation, slice thickness 5 mm, data acquisition time 22.5 seconds) was performed from the head to the mid-thigh. A PET scan was acquired immediately thereafter using an acquisition time of 3 minutes per bed position. The CT data were used for attenuation correction, and the images were reconstructed using the ordered subsets expectation maximization algorithm.
Image Analysis
Fen Zhao et al Table 1 Patient (n [ 74) and Metastatic Lesion (n [ 185) Characteristics Characteristic
Number of Patients (% or Range)
Gender Male
46 (62.16)
Female
28 (37.84)
Median age (years)
57 (38-78)
Median no. of metastatic lesions
2 (1-5)
Median no. of treatment fields
1 (1-3)
Histologic type AC
38 (51.35)
SCC
22 (29.73)
ASC
7 (9.46)
LC
7 (9.46)
Karnofsky performance score
