⁶⁸Ga-SSO-120 PET for Initial Staging of Small Cell Lung Cancer Patients: A Single-Center Retrospective Study

DISCUSSION

This is, to the best of our knowledge, the first description of SSTR2-antagonist PET imaging in SCLC patients. In the patient-based and region-based analyses, detection rates were comparable between ⁶⁸Ga-SSO-120 and ¹⁸F-FDG PET, indicating that both are valuable tools for primary staging of SCLC patients. As ¹⁸F-FDG PET is already well established, it will probably remain the mainstay of molecular imaging in SCLC patients. ⁶⁸Ga-SSO-120 PET likewise offers precise tumor detection and additional complementary information, as the region-based correlation between ⁶⁸Ga-SSO-120 and ¹⁸F-FDG uptake was low (Fig. 6; Supplemental Fig. 1). In patients with sufficient SSTR2-antagonist uptake, targeted radionuclide therapy may be performed in a theranostic approach. Of note, the cutoff of mean SUV_max (≥10) used in this work to define high uptake is not an established standard for evaluating the applicability of SSTR2-directed radionuclide therapy but was chosen to compare the intensity of the uptake interindividually and in comparison to ¹⁸F-FDG uptake. Moreover, Fendler et al. recently used an SUV_max cutoff of at least 10 in more than 50% of tumor lesions to select patients for systemic radionuclide therapy with the fibroblast activation protein inhibitor ⁹⁰Y-FAPI-46 (23).

A main goal of PET imaging in primary staging of SCLC patients is to distinguish limited disease from extensive disease to help determine the treatment (24). In this context, correct upstaging in binary VALG classification can prevent patients from undergoing ineffective surgery or radiotherapy. In this study, ⁶⁸Ga-SSO-120 PET detected more distant metastases, whereas ¹⁸F-FDG PET detected more contralateral thoracic lymph node metastases (Figs. 3 and 4). However, the additionally detected metastases did not alter VALG staging. In 2 patients, additional brain metastases on ⁶⁸Ga-SSO-120 PET were already known from cerebral MRI. Another patient had not only liver metastases that were additionally detected on ⁶⁸Ga-SSO-120 PET but also pleural and bone manifestations. In larger patient cohorts, however, recognition of additional distant metastases could potentially influence patient management regarding the decision toward curative or palliative treatment intent.

In this context, a known limitation of ¹⁸F-FDG PET in SCLC patients is detection of brain metastases, because they are barely discernible from high physiologic cerebral glucose uptake (4). ⁶⁸Ga-SSO-120 PET showed clearly detectable brain metastases in 5 patients, but in 1 patient a brain metastasis known from MRI was not detected. For 5 of these 6 patients, additional ¹⁸F-FDG PET was available (Fig. 5A), and in 2 patients, ¹⁸F-FDG uptake of brain metastases was observed. Of note, in 1 patient with several ⁶⁸Ga-SSO-120–positive brain metastases, the largest one showed no uptake but did show a discernable photopenic shape on ¹⁸F-FDG PET. In contrast, ⁶⁸Ga-SSO-120 PET is limited for detection of adrenal metastases because of the adrenal glands’ high physiologic uptake (25). In our cohort, in 2 patients adrenal metastases were detected by an irregular shape on morphologic CT and ⁶⁸Ga-SSO-120 PET images and by an inhomogeneous ⁶⁸Ga-SSO-120 uptake pattern, whereas elevated tracer uptake was difficult to evaluate (Fig. 5B).

Detection of additional thoracic lymph node metastases does not alter binary VALG staging but can, in patients with limited disease, evoke an extension of the target volume in radiotherapy planning (26). In this study, 3 patients were rated cN3 only on ¹⁸F-FDG PET, compared with 1 patient rated cN3 only on ⁶⁸Ga-SSO-120 PET. Future studies are necessary to understand the potential clinical benefit of performing a dual-tracer approach in a purely diagnostic setting. It is noteworthy that a significant number of lesions had low ⁶⁸Ga-SSO-120 uptake but were still identifiable because they had clearly increased uptake compared with the surrounding background and also exhibited suggestive morphologic features. A relevant number of these lesions yet showed high ¹⁸F-FDG uptake, with a general trend toward higher ¹⁸F-FDG uptake for primary tumor and thoracic lymph node metastases, whereas uptake was comparable in distant metastases (Fig. 6; Supplemental Fig. 1).

All patients showed any ⁶⁸Ga-SSO-120 uptake, and about 40% of patients demonstrated high ⁶⁸Ga-SSO-120 uptake (Fig. 2), with 1 patient presenting high ⁶⁸Ga-SSO-120 uptake and low ¹⁸F-FDG uptake (Fig. 7). In most patients, tumor ⁶⁸Ga-SSO-120 uptake was greater than liver uptake. We used region-based mean TLR_peak per patient as a suggested robust semiquantitative measure to evaluate lesion uptake in comparison to liver uptake. This measure was chosen on the basis of the visual Krenning score, which was originally introduced for octreotide scintigraphy and evaluates lesion uptake in comparison to physiologic reference tissue. Tumor uptake greater than liver uptake corresponds to a Krenning score of 3, and tumor uptake greater than spleen or kidney uptake corresponds to a score of 4 (27). Typically, peptide receptor radionuclide therapy can be applied to NETs if the Krenning score is at least 3. ⁶⁸Ga-SSO-120 uptake did not show a normal distribution (Supplemental Table 2), probably indicating different interindividual uptake patterns. Two previous studies used SSTR-agonist PET in SCLC patients. Both reports described enhanced tracer uptake in about half the included patients (10,11), and 1 report described any uptake in more than 80% of the evaluated lesions (11). The higher rates of patients with uptake on SSTR2-antagonist PET are in line with a previous comparison in patients with gastroenteropancreatic NETs describing an improved lesion-based detection rate for ⁶⁸Ga-SSO-120 compared with ⁶⁸Ga-DOTATOC PET, with a particular benefit for liver metastases (13).

Future evaluations of ⁶⁸Ga-SSO-120 PET in SCLC patients might point in 2 directions. First, immunohistochemical examination including determination of SSTR2 expression in lesions that have different ⁶⁸Ga-SSO-120 uptake and were previously biopsied or afterward resected might allow a deeper understanding of the various ⁶⁸Ga-SSO-120/¹⁸F-FDG uptake patterns. Most SCLC cells express SSTR2 (28), with high-level SSTR2 expression occurring in almost half of SCLC patients (29), but expression varies in SCLC subtypes with different gene signatures of transcription factors (30). If high uptake on ⁶⁸Ga-SSO-120 PET can be predicted from histopathologic examination, patients who could benefit from this examination may be selected via the initial biopsy. The other way around, ⁶⁸Ga-SSO-120 PET may serve as a noninvasive tool to guide toward specific biopsy locations (if ⁶⁸Ga-SSO-120 uptake correlates with the genetic profile, particularly in advanced disease with clonal evolution and a heterogeneous uptake pattern). For ⁶⁸Ga-DOTATATE, a correlation between histologic SSTR2 expression and both SUV_peak and TLR values was described (10). Moreover, in the increasing field of personalized medicine and theranostic options, characterization of the molecular basics of tumor biology is decisive to select the most appropriate therapy, with radionuclide therapy using SSTR2 antagonists such as ¹⁷⁷Lu-SSO-110 being a possible new theranostic approach in SCLC patients. In this context, SSTR2 PET might be a more accurate screening tool for SSTR2 positivity than immunohistochemical examination, as—regarding intraindividual heterogeneity of expression levels—it enables whole-body examination and is not affected by sampling errors.

Second, mid- and long-term follow-up of patients who underwent ⁶⁸Ga-SSO-120 PET for initial tumor staging may allow investigation of the prognostic value of ⁶⁸Ga-SSO-120 PET. Patients with sufficient ⁶⁸Ga-SSO-120 uptake might potentially benefit from SSTR2-antagonist radionuclide therapy. The first investigations at our center indicated that the fraction of patients with high ⁶⁸Ga-SSO-120 uptake may be comparable between initial staging and restaging. It will be of particular interest to investigate whether SSTR2 expression remains stable in progressive disease, because other theranostic targets are frequently lost.

Thus far, to our knowledge, no applications of SSTR2-antagonist radionuclide therapy in SCLC patients have been described. SSTR agonists have been evaluated, but the results were not sufficient for broader clinical applications. For ⁹⁰Y-DOTATOC, in 6 SCLC patients no therapy response was observed (31). In a mixed cohort of 10 SCLC and NSCLC patients who underwent ⁹⁰Y-DOTA-lanreotide treatment, a response was described in 1 patient and stable disease in 5 patients (32). In another study, ⁹⁰Y-DOTATOC/DOTATATE was applied to 7 SCLC patients and ¹⁷⁷Lu-DOTATOC/DOTATATE was applied to 4 patients, but no treatment response was observed (11). In an evaluation of 4 SCLC patients who received ¹⁷⁷Lu-DOTATATE, 1 patient had a partial response and 1 patient had stable disease (10).

At this point, SSTR2 antagonists might be beneficial because they show increased uptake (in SCLC, an antagonist-to-agonist binding ratio of 4.5 was described) and prolonged residence times (12). In NET patients, application of ¹⁷⁷Lu-SSO-110 resulted in up to 10-fold increased tumor doses and favorable tumor-to-organ dose ratios compared with ¹⁷⁷Lu-DOTATATE (15), as well as promising response rates in a prospective phase I trial (16). However, an unexpectedly high rate of hematologic toxicity was problematic but was resolved by treatment with a reduced activity and by longer intervals between treatment cycles (16). In pretreated SCLC patients, careful monitoring should be performed, but SSTR2-antagonist radionuclide therapy can be justified in settings with exhausted standard-of-care options given the limited progression-free survival and overall survival of the disease. Moreover, considering the poor survival of patients with SCLC after first-line therapy and limited options in second-line therapy, SSTR2-directed radionuclide therapy might be an option in maintenance, such as in combination with immune checkpoint inhibitors (33). A multicenter phase I trial was designed to investigate ¹⁷⁷Lu-SSO-110 in SCLC and breast cancer patients but was terminated because of a high number of screening failures (NCT03773133); however, the only SCLC patient who was screened (but was excluded because of a brain metastasis) showed very good uptake in the primary tumor and (brain) metastases on ⁶⁸Ga-SSO-120 PET. As an alternative to ¹⁷⁷Lu-SSO-110, the SSTR2-antagonist ¹⁷⁷Lu-DOTA-LM3 did not induce high-grade adverse events in NET patients (17). Moreover, ¹⁶¹Tb-labeled SSTR2 antagonists have the potential to open additional interesting theranostic opportunities due to their emitted low-energy and, thus, short-ranged β⁻-particles. In a preclinical study, ¹⁶¹Tb-DOTA-LM3 showed greater effects on survival of SSTR-positive rat pancreatic AR42J cancer cells than did ¹⁷⁷Lu-DOTA-LM3, in both cell culture and subcutaneously inoculated cells in a mouse model (34).

Patient follow-up may also be used to investigate the prognostic value of ⁶⁸Ga-SSO-120 PET in primary staging of SCLC patients. Whereas SSTR expression in SCLC was previously assumed to be associated with less aggressive tumors and potential for favoring apoptosis (5), more recent in vitro and in vivo results suggest that in this tumor entity SSTR may be a protumor survival signal (29). Consequently, in limited disease, patients with low SSTR expression showed improved survival (29). However, in a study including ⁶⁸Ga-DOTATATE PET, neither SUV metrics nor immunohistochemical scores were prognostic (10). This finding contrasts with ¹⁸F-FDG PET–derived metabolic tumor volume and total lesion glycolysis, which were prognostic of overall and progression-free survival in a recent metaanalysis (35). In this context, it is of interest that the correlation between ¹⁸F-FDG and ⁶⁸Ga-SSO-120 uptake was low (Fig. 6D; Supplemental Fig. 1D); therefore, ⁶⁸Ga-SSO-120 uptake cannot be predicted from ¹⁸F-FDG uptake.

The study faces 2 main limitations. First, the timing between ¹⁸F-FDG and ⁶⁸Ga-SSO-120 PET was heterogeneous, and some patients underwent treatment initiation between the 2 imaging modalities, potentially influencing ⁶⁸Ga-SSO-120 or ¹⁸F-FDG uptake in the case of a hypothetical very early treatment response. However, the interval was short, with a maximum of 14 d between the 2 imaging modalities, and the CT images did not show major morphologic differences between ¹⁸F-FDG and ⁶⁸Ga-SSO-120 PET (stable disease according to RECIST 1.1). Second, systemic histopathologic validation was not conducted for all lesions. However, we used a TNM classification from a clinical tumor board which was based on thoracoabdominal CT, brain MRI, whole-body PET, and endobronchial ultrasound-guided transbronchial needle aspiration to meet the highest demands of a reference standard in the setting of this retrospective analysis. Moreover, no direct comparison of the individual lesions on ⁶⁸Ga-SSO-120 and ¹⁸F-FDG PET was performed. In the planning of the study, we intentionally chose a patient- and region-based analysis, as these parameters are clinically decisive for both primary staging and evaluation of global ⁶⁸Ga-SSO-120 uptake. In radionuclide therapy planning and restaging, however, a lesion-based comparison will be of additional interest to identify patients with ⁶⁸Ga-SSO-120/¹⁸F-FDG mismatch, which might not be targeted by SSTR2-directed radionuclide therapy.