PET imaging targeting CXCR4 has the potential to distinguish different molecular subtypes of Small Cell Lung Cancer Tumors

Introduction: According to relative immunohistochemistry (IHC) expression of ASCL1, NEUROD1, and POU2F3, small cell lung cancer (SCLC) could be classified into different molecular subtypes, which has important implications for prognosis and treatment guidance. Both SCLC-A and SCLC-N subtypes are classified as neuroendocrine (NE) group, while SCLC-P and SCLC-TN (triple-negative) subtypes tend to be non-NE. In this prospective work, the feasibility of a novel PET tracer [¹⁸F] AlF-NOTA-QHY-04, targeting C-X-C-chemokine-receptor-type-4 (CXCR4) in classifying the different molecular subtypes of SCLC is verified.

Methods: 22 patients with SCLC confirmed by pathology were included in this work who underwent [¹⁸F] AlF-NOTA-QHY-04 PET/CT before any treatment. Conventional IHC staining of ASCL1, NEUROD1 and POU2F3 was performed on formalin-fixed paraffin-embedded (FFPE) samples from these 22 patients. H-score was calculated to define their molecular subtypes combined with the prominent expression of these markers, positive was defined as H-score more than 10. imaging parameters were evaluated, including maximum, mean and peak standard uptake values (SUV_max, SUV_mean and SUV_peak). Tumor-to-normal ratios of tracer uptake tissues based on SUV_max were further calculated, denoted as T/NT (T/NT_lung and T/NT_blood). Data for SUV_max, SUV_mean, SUV_peak, T/NT_blood and T/NT_lung were expressed as mean ± standard deviation (SD). Two-sample t tests and Mann-Whitney u tests were used to compare the PET/CT parameters between different subtypes. The diagnostic value of these imaging parameters was analyzed between NE vs non-NE groups using receiver operating characteristics (ROC) analysis and area under the curve (AUC).

Results: According to the prominent IHC expression in the 22 patients’ FFPE specimens, the molecular subtypes of SCLC were SCLC-A (n=13, 59.1%), SCLC-N (n=2, 9.1%), SCLC-P (n=4, 18.2%), and SCLC-TN (n=3, 13.6%). Parameters including SUV_max, SUV_mean, SUV_peak, T/NT_blood and T/NT_lung of primary lesions didn’t show significant difference among these four molecular subtypes. In NE group, SUV_mean, T/NT_blood and T/NT_lung of lymph node metatases were significantly higher in SCLC-N than SCLC-A (all P<0.05), no difference in non-NE group. However, SUV_max, SUV_mean, SUV_peak, T/NT_blood and T/NT_lung of lymph node metatases were all significantly higher in SCLC-N than SCLC-P (all P<0.05), and SUV_max, T/NT_blood and T/NT_lung all significantly higher in SCLC-N than SCLC-TN (all P<0.05). Parameters of primary tumor were found higher in NE group than non-NE group such as SUV_max (8.165±2.289 v.s 6.340±1.294, P=0.0251) and T/NT_lung (12.900±4.028 v.s 9.026±1.980, P=0.0268). In lymph node metastases, no difference was observed in all parameters between NE and non-NE groups. The ROC analysis demonstrated that the cutoff value of SUV_max (AUC=0.800) and T/NT_lung (AUC=0.819) of biopsy site were 6.28 and 8.625. These parameters had high sensitivity, specificity and accuracy in distinguishing non-NE and NE group.

Conclusions: Our preliminary findings indicated that [¹⁸F] AlF-NOTA-QHY-04 PET/CT imaging might be a useful and non-invasive tool for distinguishing different molecular subtypes of SCLC.