Visual Abstract
Abstract
Posttreatment imaging of γ-emissions after peptide receptor radionuclide therapy (PRRT) can be used to perform quantitative dosimetry as well as assessment response using qualitative measures. We aimed to assess the impact of qualitative posttreatment imaging on the management of patients undergoing PRRT. Methods: In this retrospective study, we evaluated 100 patients with advanced well-differentiated neuroendocrine tumors undergoing PRRT, who had posttreatment SPECT/CT imaging at 24 h. First, we evaluated the qualitative assessment of response at each cycle. Then using a chart review, we determined the impact on management from the posttreatment imaging. The changes in management were categorized as major or minor, and the cycles at which these changes occurred were noted. Additionally, tumor grade was also evaluated. Results: Of the 100 sequential patients reviewed, most (80% after cycle 2, 79% after cycle 3, and 73% after cycle 4) showed qualitatively stable disease during PRRT. Management changes were observed in 27% (n = 27) of patients; 78% of those (n = 21) were major, and 30% (n = 9) were minor. Most treatment changes occurred after cycle 2 (33% major, 67% minor) and cycle 3 (62% major, 33% minor). Higher tumor grade correlated with increased rate of changes in management (P = 0.006). Conclusion: In this retrospective study, qualitative analysis of posttreatment SPECT/CT imaging informed changes in management in 27% of patients. Patients with higher-grade tumors had a higher rate of change in management, and most of the management changes occurred after cycles 2 and 3. Incorporating posttreatment imaging into standard PRRT workflows could potentially enhance patient management.
Neuroendocrine neoplasms encompass a diverse range of tumors, primarily originating in the lungs and gastroenteropancreatic sites. The classification of gastroenteropancreatic neuroendocrine neoplasms by the World Health Organization into different histopathologic subgroups provides prognostic insights based primarily on differentiation status, mitotic rate, or Ki-67 proliferation index; well-differentiated tumors are subclassified into G1 (Ki-67, <3%), G2 (Ki-67, 3%–20%), and G3 (Ki-67, >20%) neuroendocrine tumors (NETs), which exhibit increasing aggressiveness as the proliferation rate increases; poorly differentiated neuroendocrine carcinomas represent the most aggressive subset and include large and small cell subtypes (1). Despite the relatively low incidence of each individual subtype, the global burden of neuroendocrine neoplasms is on the rise (2,3).
Peptide receptor radionuclide therapy (PRRT) using 177Lu-DOTATATE has become a standard treatment for well-differentiated NETs (4–8) and is given clinically as 4 cycles of 7.4 GBq every 8 wk. However, with increased clinical experience, adapting standard treatment protocols to specific patients has become more common. For example, in patients who experience bone marrow toxicity, administered activity can be decreased or the interval between PRRT treatments can be increased (9).
Although somatostatin receptor (SSTR) PET is used to select patients and follow patients after treatment (10), there are currently no guidelines for how to use posttreatment imaging of the administered 177Lu-DOTATATE (referred to as posttreatment imaging here) during PRRT, and most treatment centers in the United States do not routinely perform immediate posttreatment imaging. Similarly, the latest published guidelines from the North American Neuroendocrine Tumor Society and the Society of Nuclear Medicine and Molecular Imaging regarding administration of 177Lu-DOTATATE PRRT do not specifically address the utility of performing posttreatment imaging (4).
Posttreatment imaging can be used in 2 major ways: the first is to perform quantitative dosimetry of tumors and normal organs, and the second is to qualitatively evaluate the patient’s response to treatment. Quantitative dosimetry has been shown to correlate with radiographic response in gastroenteropancreatic NETs but not overall survival (11,12). Although there is a relationship between lesion absorbed dose and response, it is unclear how to apply this quantitative data to change management on a patient level. In terms of the role of qualitative interpretation of posttreatment imaging, there is no evidence of its current role or value. A study evaluating the practice of radiopharmaceutical therapies across European countries revealed that each one of them performed posttreatment imaging for 177Lu-DOTATATE PRRT (13). The study highlighted that the impact of posttreatment imaging on clinical decision-making remains uncertain.
The role of qualitative posttreatment imaging–based response and its impact on management is not clearly understood, and therefore, we aimed to investigate the impact of posttreatment imaging on changes in management in patients undergoing 177Lu-DOTATATE PRRT at our institution.
MATERIALS AND METHODS
Study Population
We conducted a retrospective study of the first 107 patients who underwent 177Lu-DOTATATE PRRT for well-differentiated NETs at our institution between 2018 and 2020. The patient cohort included individuals who had received a minimum of 2 cycles of PRRT as per standard treatment guidelines (4). Anatomic imaging was performed as a part of our institutional protocol after cycle 2, and characterization of pseudoprogression was based on these images. The institutional review board approved this retrospective study, and the requirement to obtain informed consent was waived.
Posttherapy Scan Acquisition
Whole-body planar and SPECT/CT imaging were performed 1 d after each cycle in the context of routine clinical care using a dual-head γ-camera Infinia Hawkeye (GE Healthcare) system with the following acquisition parameters: 208% ± 10% keV photopeak, 170% ± 10% keV scatter window, 128 × 128 matrix, 30 s per projection, 60 projections in total using 2 detectors, medium-energy general-purpose collimators, and a low-dose CT for attenuation correction. A Xeleris workstation (GE Healthcare) was used for reconstruction with the following reconstruction parameters: ordered-subset expectation maximization, 10 iterations, 6 subsets, and a Butterworth filter, with scatter correction and attenuation correction. The imaging duration was approximately 60 min, consisting of 1 whole-body planar acquisition and two 20-min SPECT/CT bed positions covering the kidneys and most of the tumor.
Posttherapy Scan Analysis
Consecutive posttreatment scans were compared with the baseline scan performed after cycle 1. Response was qualitatively assessed using both SPECT/CT and planar images and was divided into 4 subtypes: marked reduction in tumor volume, reduction in tumor volume but with significant residual disease, stable disease, and development of new lesions. After each cycle, the clinical management for each patient was evaluated to determine whether the posttreatment imaging influenced the treatment plan. Patients whose posttreatment imaging resulted in a change in their management were further analyzed to identify the extent of the impact, which was broken down into either major or minor changes. Major changes included early discontinuation of PRRT (before completion of 4 cycles); for example, stoppage of PRRT because of progressive disease or because of marked response, delay or deferral of PRRT with a recommendation for targeted treatment of a new or growing lesion based on posttherapy imaging, or stoppage of PRRT because of developing hematologic toxicities with substantial imaging response. Minor changes included changes that resulted in tailoring of PRRT cycles to specific occurrences; for example, continuation of PRRT despite development of borderline hematologic toxicities, pseudoprogression, or hydronephrosis leading to stent placement. Additionally, we divided these changes into subgroups (objective vs. subjective changes), with objective changes indicating management changes that would be considered within the standard of practice.
Statistical Analysis
Descriptive statistics in the form of mean, median, SD, and ranges were used to describe quantitative variables from the clinical data. Categoric variables were reported as counts and percentages. A Pearson χ2 test was conducted to assess the relationship between the grade of tumor and the rate of changes in management at a predetermined significance level of less than 0.05. If more than one type of change in management was noted in a single patient, major changes usurped minor changes and objective changes usurped subjective changes for analysis.
RESULTS
Patient Characteristics
In total, 107 sequential patients with well-differentiated NETs underwent PRRT from May 2016 to April 2021. Seven patients did not receive posttreatment imaging and were not included in the analysis. Patient and demographic data are provided in Table 1.
Patient Demographic and Tumor Features or Patient Clinical and Demographic Features
Qualitative Response Assessment
All 100 patients had posttreatment imaging performed after cycles 1 and 2, 85% of patients had imaging after cycle 3, and 64% of patients had imaging after cycle 4. The most common imaging response was stable disease, which was seen in 73%–80% of the posttreatment images depending on the cycle. Ten patients had a marked response seen on posttreatment imaging (Table 2).
Qualitative Response Assessment on Posttreatment Imaging After Each PRRT Cycle
Change in Management Based on Posttreatment Imaging
Posttreatment imaging resulted in a change in management in 27% (27/100) of patients, with 78% (21/27) experiencing a major change and 33% (9/27) having a minor change. When broken down on the basis of subjective or objective changes in management, 59% (16/27) experienced an objective change and 44% (12/27) experienced a subjective change in management. Within the 27 patients, 30 major or minor management changes were observed, with 3 patients experiencing multiple changes in management (Table 3).
Types of Change in Management Based on Posttreatment Imaging and Cycles After Which Change Was Noted
Among the 21 patients exhibiting major changes, 3 had further PRRT cycles stopped because disease progression was found on posttreatment imaging, 5 had discontinuation of further cycles because of a marked response to PRRT, 6 had subsequent PRRT cycles delayed to allow for targeted treatment of a new or growing lesion, and 7 had further PRRT cycles deferred because of hematologic toxicity in the setting of imaging response to PRRT (Table 3; Supplemental Tables 1 and 2; supplemental materials are available at http://jnm.snmjournals.org). Among the patients with major changes, the change was made after cycle 2 for 7 patients (33%), after cycle 3 for 13 patients (62%), and after cycle 4 for 1 patient (5%) (Fig. 1).
Impact of posttreatment imaging on management, broken down by cycle when change in management occurred.
Of the 9 patients who had minor changes, 2 had PRRT continued for 4 cycles despite borderline hematologic toxicities, 1 had a stent placed for hydronephrosis, and 6 had pseudoprogression detected on conventional imaging and characterized on posttreatment imaging (Table 3). Among the patients with minor changes, the change was made after cycle 2 in 6 patients (67%) and after cycle 3 in 3 patients (33%); no minor change in management was noted after cycle 4 (Fig. 1).
In 3 patients, both major and minor changes were noted: in 2 patients, pseudoprogression was characterized after cycle 2 and treatment was stopped early because of a marked response after cycle 3; in 1 patient, treatment was stopped in the setting of developing hematologic toxicity with partial response, and a renal stent was placed because of marked hydronephrosis.
Tumor Grade and Change in Management
Patients with a higher tumor grade had a higher rate of change in management. A significant relationship was noted between the tumor grade and the change in management. Of the patients with grade 1 and 2 tumors, 23 of 95 had a change in management, whereas in patients with grade 3 tumors, 4 of 5 patients had a change in management (P = 0.006; Fig. 2).
Change in management based on grade of tumor.
Case Examples
Six cases are provided to highlight how posttreatment imaging results in changes in management.
The first case demonstrates a patient with a marked response. This patient was an 80-y-old man with a grade 2 pancreatic NET for whom treatment was stopped after 2 cycles because of marked response. His disease progressed 41 mo after the initiation of PRRT (Fig. 3).
80-y-old man with grade 2 small-bowel NET treated with 2 cycles of 177Lu-DOTATATE, demonstrating major change in management with stopping of PPRT in setting of marked response. (A) Pretreatment 68Ga-DOTATATE PET demonstrates SSTR-positive disease (arrow). (B) Postcycle 1 planar imaging demonstrates uptake in osseous and hepatic disease (arrow). (C) Postcycle 2 planar imaging demonstrates reduction in uptake in previously visualized disease (arrow). Treatment was stopped because of marked response, and disease progressed after 41 mo since start of treatment. (D) 68Ga-DOTATATE PET demonstrates no evidence of progression after 12 mo of PRRT. (E) 64Cu-DOTATATE PET demonstrates SSTR-positive disease on progression; however, tumor volume decreased compared with baseline.
The second case demonstrates a patient with progressive disease. This patient was a 43-y-old man with a grade 3 pancreatic NET for whom treatment was stopped after 3 cycles because of evidence of disease progression on posttreatment imaging. The patient was converted to treatment with chemotherapy (Fig. 4).
43-y-old man with grade 3 pancreatic NET (Ki-67, 40%) treated with 3 cycles of 177Lu-DOTATATE, demonstrating major change in management with stopping of PRRT because of progressive disease on posttreatment imaging. (A) Pretreatment 68Ga-DOTATATE PET demonstrates liver-dominant disease. (B) Postcycle 1 planar imaging demonstrates uptake in nodal and hepatic disease (arrow). (C) Postcycle 3 planar imaging demonstrates mixed response to treatment, with increase in SSTR-avid tumor volume in midline (arrow) but slight reduction in right lobe liver disease (dotted arrowhead). Although there was partial response in some lesions, cycle 4 was not administered because of evidence of progression, and patient was converted to treatment with chemotherapy.
The third case demonstrates a patient with a major change in management due to a new lesion on posttreatment imaging. This was a 68-y-old woman with a grade 2 pancreatic NET in which a new lesion was detected on the L4 vertebra on imaging after cycle 4. The patient was given stereotactic body radiation therapy for the new vertebral lesion immediately after PRRT (Fig. 5).
68-y-old woman with grade 2 pancreatic NET treated with 4 cycles of 177Lu-DOTATATE, demonstrating major change in management with treatment of new lesion detected on posttreatment imaging. Pretreatment 68Ga-DOTATATE PET/CT (A, fused SPECT/CT [top] and CT from PET/CT [bottom]) and postcycle 1 images (B, whole-body planar image [left], fused SPECT/CT [top], CT from SPECT/CT [bottom]) demonstrate no evidence of lesion in L4 vertebra (dashed circles). (C, whole-body planar image [left], fused SPECT/CT [top], and CT from SPECT/CT [bottom]) Postcycle 4 planar and SPECT/CT imaging demonstrates uptake in L4 vertebrae with no corresponding CT abnormality (white and black arrows). Patient developed back pain at cycle 4, and MRI demonstrates new lesion (D, arrow). Patient was treated with stereotactic body radiation therapy to L4 lesion immediately after cycle 4.
The fourth case demonstrates a patient for whom treatment was halted after cycle 2 to treat a SSTR-negative lesion. This was a 76-y-old man with a grade 2 bronchial carcinoid who had a growing SSTR-negative hepatic lesion after cycle 2. The patient underwent transarterial chemoembolization for this lesion before resuming cycle 3 of PRRT (Fig. 6).
78-y-old man with grade 2 bronchial carcinoid treated with 3 cycles of 177Lu-DOTATATE, demonstrating major change with delay of cycle 3 to manage growing SSTR-negative hepatic lesion after cycle 2. Pretreatment CT (A, dotted arrow) and 68Ga-DOTATATE PET (D) done within month of cycle 1 of PRRT demonstrate no evidence of hepatic lesion. CT imaging done after cycle 2 of treatment (B) demonstrates lesion in segment VI of liver (arrow). Postcycle 2 imaging (C, dotted circle) demonstrates no uptake in this hepatic lesion, although posttreatment 68Ga-DOTATATE PET from cycle 1 to cycle 3 demonstrates reduction in SSTR-positive chest wall disease (E–G, arrow). Cycle 4 was abandoned because patient had worsening of clinical symptoms.
The fifth case demonstrates a patient with minor change in management with characterization of pseudoprogression after cycle 2. This was a 68-y-old woman with a grade 1 small-bowel NET treated with 4 cycles of 177Lu-DOTATATE, in which a growing hepatic lesion on MRI was characterized as pseudoprogression on posttreatment imaging at cycle 2 (Fig. 7).
68-y-old woman with grade 1 small-bowel NET (Ki-67, 1%) treated with 4 cycles of 177Lu-DOTATATE, demonstrating minor change in management with characterization of pseudoprogression after cycle 2. Pretreatment MRI (A) done 3 wk before cycle 1 demonstrates hepatic lesion with increase in size on subsequent MRI (B) done after cycle 2 of PRRT (arrow). Growth was characterized as pseudoprogression on posttreatment SPECT/CT after cycle 2 given unchanged appearance from cycle 1 (D and E, dotted arrows). Posttreatment SPECT/CT after cycles 3 and 4 (F and G, dotted arrows) demonstrates reduction in uptake of lesion from cycle 1 to cycle 4. Postcycle 4 MRI demonstrates reduction in size of hepatic lesion (C, arrow).
The sixth case demonstrates a patient for whom PRRT was stopped because of developing hematologic toxicities in the setting of imaging response. This was a 60-y-old man with a grade 3 pancreatic NET who had a good response to PRRT except for bulky pancreatic disease after cycle 3. There was an impending risk of hematotoxicity with grade 2 toxicity per the Common Terminology Criteria for Adverse Events, and PRRT was discontinued given the partial response on posttreatment imaging. SSTR-positive pancreatic disease was subsequently managed surgically after his liver disease was stable over a period of 3 mo after PRRT cycles (Supplemental Fig. 1)
DISCUSSION
We demonstrated that 27% of patients with NETs being treated with PRRT underwent a change in management based on posttreatment SPECT/CT imaging. Patients with higher-grade tumors had a higher rate of change in management. Most of the changes in management occurred after cycles 2 and 3.
To our knowledge, this is the first description of the qualitative impact of posttreatment SPECT/CT imaging on the management of patients undergoing PRRT. Although most prior work has focused on the role of quantitative dosimetry for the management of patients, there is a growing interest in the qualitative impact of posttreatment scans as well. A previous study found that thoracoabdominal SPECT/CT imaging is the preferred method for post-PRRT imaging and that all accompanying CT images should be reviewed for additional findings, such as ascites (14). Although our study builds on the existing body of literature, our results demonstrate that posttreatment SPECT/CT imaging can have significant value as a qualitative marker of response, which can directly impact patient management.
Current guidelines do not include recommendations for performing posttreatment imaging, primarily because there is currently little direct impact of quantitative dosimetry on patient management. Our results demonstrate that qualitative posttreatment imaging is critical to patient management. Baseline imaging after cycle 1 should be obtained for comparison of future posttreatment images. Given that most changes in management occurred after cycles 2 and 3, images should be considered after these 2 cycles. The role of postcycle 4 images is unclear, but the images may be valuable for comparison to future imaging studies in patients. On the basis of our analysis, it is evident that for grade 3 tumors, posttreatment imaging should be performed after each cycle because these patients exhibit a high rate of change in management.
Although our practice is to stop PRRT early in patients with marked response, it is unclear if this is the appropriate way to manage patients. The average time to disease progression after PRRT in the subgroup in which PRRT was halted early was 27 mo. These results suggest that it is safe to stop treatment early and save the remaining cycles of treatment for subsequent use; however, further research is warranted to understand the appropriate management.
In the future, we hope that quantitative dosimetry will lead to patient-specific adjustments in treatment. There is currently extensive interest in understanding and using quantitative dosimetry in radioligand therapies (11,12,15–17). Obtaining posttreatment imaging for qualitative assessment allows for the accumulation of data that can be later used to understand the role of quantitative dosimetry. Although the absorbed dose in grays can be used for response assessment, our result suggests that qualitative evaluation has an important impact on patient management, and therefore, other quantitative measures such as SPECT-based SUVs may be useful as response markers without needing to be converted to grays. Although we did not evaluate SUV in our paper, our results suggest that this quantitative approach, rather than a dosimetric quantitative approach, may be a valuable tool for evaluation response to PRRT in the future. Currently, significant ongoing work implementing SPECT-based quantitative uptake will enable this approach in the future (18,19). An important consideration while assessing response with SUV is to consider the tumor sink effect and its impact on SUV measurements as patients start responding to PRRT (20).
There is controversy about when to acquire conventional imaging (CT and MRI) or functional imaging (SSTR PET) during and after PRRT. Currently, anatomic imaging remains the backbone for disease assessment, but anatomic imaging often lags functional imaging response. We identified a subgroup of patients who demonstrated progression on structural imaging but were characterized as pseudoprogression on posttreatment imaging, which suggests that posttreatment imaging may be a better approach for response assessment than anatomic imaging. Additionally, posttreatment SPECT may be able to replace SSTR PET for response evaluation. Having said this, how conventional imaging and posttreatment imaging complement each other remains an open question. Moving forward, we hope that guidelines will embrace posttreatment imaging because of its impact on patient management, as we continue to learn how to leverage the qualitative dosimetry data acquired on the images. SPECT/CT scanners are widely available in nuclear medicine departments, and most centers offering PRRT would have access to SPECT/CT scanners.
There are many limitations with this study, in particular, the small cohort of patients in which a change in management based on posttreatment imaging was noted. Second, this study is retrospective in nature, and prospective evaluation would help validate the results. Third, in the subgroup in which PRRT was stopped early because of impending hematotoxity (Common Terminology Criteria for Adverse Events grade 2 and above) in the setting of stable or partial response on posttreatment imaging, the contribution of posttreatment scans alone remains unclear because of multiple factors causing early discontinuation of PRRT. Lastly, performing mid-cycle CT or MRI is not yet standardized, and characterization of pseudoprogression may not be relevant at other institutions where these scans are not routinely performed. The cost implications and frequent visits associated with posttreatment imaging are worth considering in this context. However, a point to note is that 4 SPECT/CT scans are approximately the same cost as 1 SSTR PET scan. Additionally, we chose the 24-h time point because it is convenient for patients who might be traveling for PRRT, not to mention the benefit of not undergoing additional radionuclide injection.
CONCLUSION
In this retrospective study, qualitative analysis of posttreatment SPECT/CT led to changes in management in 27% of patients. Patients with higher-grade tumors had a higher rate of change in management, and most of the changes in management occurred after cycles 2 and 3. It may be valuable to incorporate posttreatment imaging in the standard PRRT workflow to aid in patient management.
DISCLOSURE
Thomas Hope has grant funding to the institution from Clovis Oncology, GE Healthcare, Lantheus, Janssen, the Prostate Cancer Foundation, Telix, and the National Cancer Institute (R01CA235741 and R01CA212148). He received personal fees from Bayer, BlueEarth Diagnostics, and Lantheus and received fees from and has an equity interest in RayzeBio and Curium. No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: What is the clinical role of posttreatment imaging during PRRT?
PERTINENT FINDINGS: Qualitative posttreatment imaging triggered a change in management in 27% of patients undergoing PRRT, with most changes occurring after cycle 2 (37%) and cycle 3 (59%). Higher tumor grade was associated with a higher rate of change in management.
IMPLICATIONS FOR PATIENT CARE: Incorporating posttreatment SPECT/CT imaging into standard PRRT workflows could potentially facilitate personalized patient management.
- © 2024 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication September 5, 2023.
- Accepted for publication December 20, 2023.
