Thesis title: Clinical impact of different 18F-FDG PET/CT criteria for the evaluation of immunotherapy response in advanced melanoma.
The introduction in the clinical practice of anti CTLA-4 and anti-PD-1 immune checkpoint inhibitors (ICIs) and target therapy (BRAF and MEK inhibitors) has radically modified the therapeutic perspective in patient affected by advanced melanoma, since these new drugs allowed to increase the average OS from the historic 6-9 months to about 3 years.
The efficacy of ICIs has been demonstrated against different types of malignancies, including lymphomas, melanoma, renal cell carcinoma and non-small cell lung cancer (NSCLC); however, it is important to remark that immunotherapy with ICIs is effective only for a part of treated patients, about 20-40%.
It is therefore essential to differentiate responders from non-responders as soon as possible; since immunotherapy with ICIs acts on molecular pathways completely different from chemotherapy, the assessment of treatment response with traditional radiological approaches has several limitations.
In fact, the key point of the anti-neoplastic effect of these drugs is the recruitment and activation of immune system cells in cancer lesions, which can expand before regress, or may remain metabolically stable for a long time. Therefore, the evaluation of immunotherapy response with diagnostic imaging requires a different approach than the one used for conventional anti-tumor therapies.
Regarding the assessment of immunotherapy response, the role of 18F-FDG PET/CT seems to be very promising, but has yet to be fully defined, considering that the conventional criteria for evaluation of treatment response (EORTC, PERCIST) have different limitations in terms of diagnostic accuracy.
Therefore, in recent years different assessment criteria have been introduced, based on the combination of functional and biological data (PECRIT) or on the assessment of the absolute number of newly emergent lesions and their functional dimensions (PERCIMT), with the aim of improving the diagnostic performance of PET/CT. The purpose of this study is to compare the ability of different PET evaluation criteria (PERCIST vs PERCIMT) to predict the six-month response (i.e. best overall response at 6 month) to immunotherapy with ICIs in patients affected by metastatic melanoma, and to identify possible differences in the diagnostic accuracy of the different methods.
Materials and methods
From December 2018, we retrospectively evaluated a cohort of 23 patients affected by advanced melanoma, treated with ICIs and submitted to 18F-FDG PET/CT scans performed in our Nuclear Medicine Unit.
The enrolled patients underwent a 18F-FDG PET/CT scan before the start of ICIs (SCAN 1), and a control scan (SCAN 2) at the end of therapy with Ipilimumab or after an average of 12 weeks from the start of Nivolumab/Pembrolizumab.
The treatment response was evaluated by comparing the results of SCAN 2 with those of SCAN 1 using both PET Response Evaluation Criteria in Solid Tumors (PERCIST) and PET Response Evaluation Criteria for Immunotherapy (PERCIMT).
The scans interpreted as "pseudo-progression" at SCAN 2 according to the PERCIMT criteria were subjected to a new 18F-FDG PET/CT confirmatory control (SCAN 3) 4-6 weeks later; the patients in which SCAN 2, according with PERCIMT criteria, revealed disease progression or the onset of serious irAE, have been subjected to a therapeutic switch to a different line of immunotherapy or target therapy.
The "best overall response" (BOR) was determined for each patient by the oncologist at about 6 months from the start of treatment on the basis of clinical examination, followed by subsequent 18F-FDG PET/CT confirmatory studies or different imaging modalities (contrast enhancement CT or MRI).
The statistical analysis was performed through the construction of receiver operating characteristic (ROC) curves, in order to estimate the predictive value of each PET derived criteria on patient’s response to immunotherapy. For each patient, the time difference (in months) between staging PET and the need to change treatment following the onset of progression disease at follow-up PET controls (time to switch) was assessed, measured according to both PERCIST and PERCIMT criteria. The student’s T test was performed to assess the statistical significance of the difference in time to switch according to the criteria adopted (PERCIST vs PERCIMT), fixing the significance to P < 0.05.
Results
Among the analyzed criteria, the PERCIMT criteria showed the highest predictivity of BOR at 6 months. ROC curve analysis for predicting the immunotherapy response for each PET-derived criteria showed that PERCIST and PERCIMT predicted response with an area under the curve respectively of 0.7 (sensitivity 90,9%, specificity 50%, accuracy 69.5, P = 0.02) and 0.87 (sensitivity 90.9%, specificity 83.3%, accuracy 86.9, P < 0.001). The statistical analysis of the results performed using student’s T test found an average TTS of 3.2 months (SD 0.46) for PERCIST criteria, compared to 5.3 months (SD 3.58) for PERCIMT criteria, with a statistically significant difference between the two criteria (P = 0.04).
Conclusions
The 18F-FDG PET/CT scan evaluation performed according to PERCIMT criteria significantly improves the performance in terms of diagnostic accuracy compared to PERCIST criteria, mainly in terms of specificity.
However, considering the sensitivity values emerged from our study, since in presence of cases classified as pseudo-progression according to PERCIMT criteria it is not possible to certain exclude a real disease progression, currently the most prudent approach is to perform a short-term control scan, in order to confirm whether treatment is effective or to identify non-responders at an early stage. Since immunotherapy is not the only therapeutic option available for patients affected by advanced melanoma, it is important to stress that the dynamic use of different criteria, based on the type of treatment at the time of the scan, allows to better guide the clinician in the management of the patient; whereas it is well known that the metabolic information resulting from the use of PET/CT in the assessment of treatment response allows clinicians in general to identify disease progression at an early stage compared to the morphological criteria of the conventional radiological imaging techniques, in the specific case of immunotherapy, the dynamic use of the different PET criteria available, allows at the same time to avoid too premature therapeutic switches, and to obtain certain advantages in terms of PFS; the results of our study seem to confirm this topic, but should be supported by long-term survival data. Finally, our study confirms the important role of 18F-FDG PET/CT in early detecting immune-related adverse effects, compared to traditional radiological imaging techniques, the careful management of which contributes to improve the outcome of the patient.