Early-time-point ¹⁸F-FDG-PET/CT and other prognostic biomarkers of survival in metastatic melanoma patients receiving immunotherapy

We enrolled patients 18 years of age or older who had histologically confirmed, unresectable, advanced melanoma and were planned to be treated per standard of care with ICIs with anti-cytotoxic T-lymphocyte-associated antigen 4 (anti-CTLA-4) and/or anti-programmed death-1 (anti-PD-1) treatment in the first or second line of systemic treatment at the Institute of Oncology Ljubljana, Slovenia. The key exclusion criteria included symptomatic brain metastases and malignant diseases other than melanoma.

In this noninterventional, prospective study, patients underwent baseline ¹⁸F-FDG PET/CT within four weeks before treatment initiation and were monitored regularly with serial ¹⁸F-FDG PET/CT: at week four (W4) (+/- 5 days), week 16 (W16) (+/- 7 days), and week 32 (W32) (+/- 7 days) after treatment initiation and every 16 weeks thereafter. The first follow-up ¹⁸F-FDG PET/CT at W4 was performed for investigational purposes and was not necessarily used to guide treatment decisions. The clinical data and images included in this analysis were obtained from disease diagnosis up to 1^st September 2024.

All ¹⁸F-FDG PET/CT data were acquired before and during ICI treatment, and all clinical data were collected for review. The irAE grade was assigned prospectively and scored with the use of the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE, v.5.0).¹³ The irAE were classified as serous irAE in case of higher grade of irAE of 3 and above. Imaging and clinical data were anonymized and stored in a secure LabKey database server.¹⁴

The clinical protocol was approved by the Ethics Committee ERIDEK-0034/2020 and the Clinical Trials Protocol Review Committee ERID-KSOPKR-0032/2020 at the Institute of Oncology Ljubljana and by the Commission of the Republic of Slovenia for Medical Ethics (approval number: 0120-256/2020-14, September 15^th 2020). It was conducted following the ethical standards defined by the Declaration of Helsinki and the International Conference on Harmonization Guidelines for Good Clinical Practice. The study was registered with ClinicalTrials.gov under the registration number NCT06207747.

The study was conducted with the acknowledgment and consent of the subjects. All patients provided signed informed consent for treatment and consent allowing the use of their data for scientific purposes.

All ¹⁸F-FDG PET/CT scans were obtained on Biograph mCT PET/CT (Siemens, Knoxville, TN). Imaging protocol required patients to fast for 6 hours prior to injection of the radiotracer and have a blood glucose level below 10 mmol/L at the time of the scan. Patients were required to hold all diabetic medication, including metformin, for 6 hours prior to radiotracer injection. All scans were acquired per standard of care. CT that meets response evaluation criteria in solid tumours (RECIST) analysis needs was acquired according to adjusted protocol including sinogram affirmed iterative reconstruction (SAFIR) to minimize dose. Following reconstruction, PET images were normalized by patient weight and injected dose to compute SUV. More details about image acquisition can be found in our previous paper, where the same cohort of patients was used for analysis.¹¹

The tumor response to ICIs on ¹⁸F-FDG PET/CT was evaluated by a nuclear medicine specialist combining the European Organisation for Research and Treatment of Cancer (EORTC) criteria and visual response assessment.¹⁵ The SUV_max and size of the lesions were measured in all most representative tumor lesions, which are the largest lesions of a certain area or organ with the highest FDG uptake at baseline, W4 and all consecutive ¹⁸F-FDG PET/CT scans. Patients were classified into four major categories on the basis of the tumor response to ICIs: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD) and progressive metabolic disease (PMD).¹⁵ According to the information gathered after the whole -body ¹⁸F-FDG PET/CT visual assessment, two new categories, of heterogenous response (HGR) and possible pseudo-progressive disease (PPD), were added for tumor response evaluation. HGR was assigned when multiple lesions were variably meeting the criteria of PMD, SMD, PMR and CMR and could not be classified into only one response evaluation category. PPD was assigned in the case of moderate metabolic progression of the baseline tumor lesions with few locally distributed new lesions. Obvious progression with multiple new tumor lesion sites was classified as true progression (PMD). The patients were further stratified into two groups: patients with PMD were classified into the no clinical benefit (no-CB) group and patients with other response categories into the clinical benefit (CB) group. A summary of the response criteria is presented in Supplementary Table 1.

The primary end point of this study was the analysis of median OS based on the W4 ¹⁸F-FDG PET/CT response. The secondary endpoint was the median OS, which was analysed on the basis of the level of LDH, the number of organs with metastatic involvement at the beginning of ICI treatment, occurrence of irAE, higher irAE, cutaneous irAEs, endocrine irAEs and immune-related thyroiditis (irThyroiditis).

Patient characteristics were summarized via descriptive statistics. Survival analysis was performed via the Kaplan–Meier method, and 95% confidence intervals (CIs) were calculated. The associations of each of the eight metrics with OS were assessed with a univariate Cox proportional hazard model. With the use of the Bonferroni correction for testing multiple hypothesis, probability values p < 0.006 were considered statistically significant.

The Kaplan–Meier estimated OS and progressionfree survival (PFS) for the whole patient group were 18.5 months (95% CI = 14.4 months – not reached [NR]) and 8.1 months (95% CI = 4.3–26.3 months), respectively (Figure 1). Among the whole group, 39 (55%) patients died, and 44 (62%) patients progressed to immunotherapy before the cut-off date. On Figure 2, the swimmer plot displays data for individual patients, where each horizontal line or bar shows type and duration of treatment, and each point represents either ¹⁸F-FDG-PET/CT or end-of-study reason (if applicable).

FIGURE 1.

Kaplan-Meier curves show overall survival (OS) probability (A) and progression-free survival (PFS) probability (B) over time in patients with metastatic melanoma treated with immune checkpoint inhibitors. The grey shading reflects the 95% confidence interval.

FIGURE 2.

Swimmer plot shows individual patient’s treatment progression in each horizontal line. Colourful bars and lines indicate type and duration of treatment, while dots indicate specific action - ¹⁸F-FDG-PET/CT imaging or reason for end-of-study (if applicable). Vertical dashed lines indicate a time when ¹⁸F-FDG-PET/CT scan should be performed for patients in this study.

Among the 68 (96%) patients who underwent W4 ¹⁸F-FDG PET/CT, 51 (72%) patients were classified into the CB group, and 17 (24%) were classified into the no-CB group. The median OS was not reached (NR) (95% CI = 17.8 months - NR) in the CB group and was 6.2 months (95% CI = 4.6 months - NR) in the no-CB group (Figure 3). In univariate Cox analysis classification was statistically significantly correlated to OS with hazard ratio (HR) of 0.4 (95% CI = 0.18–0.72; p = 0.004).

FIGURE 3.

Kaplan-Meier curves showing probability of median overall survival (OS) between clinical benefit (CB) and no-CB group as classified by findings on week four (W4) ¹⁸F-FDG PET/CT. The curves are statistically significantly different (p = 0.03).

Among the 17 patients with PMD, who were classified into the no-CB group, 7 (42%) died before the W16 ¹⁸F-FDG PET/CT scan. Three (18%) patients had PMD on W16 ¹⁸F-FDG PET/CT, 6 (35%) had PMR, and one (6%) patient had CMR. Two (12%) patients with PMD on W4 ¹⁸F-FDG PET/CT changed systemic therapy from ICI therapy to targeted therapy because of clinical and radiological signs of rapid progression affecting vital organs.

Three patients were classified as CMR on W4 ¹⁸F-FDG PET/CT. One patient achieved a durable response with CMR, one patient experienced fatal grade 5 immune-related encephalitis during treatment and one experienced local progression in soft tissues 25 months after CMR imaging on W4 ¹⁸F-FDG PET/CT. The site of progression was amenable for local treatment with radiotherapy, and a complete metabolic response was achieved.

In Figure 4, the alluvial plot shows the responses on ¹⁸F-FDG PET/CT scans for each patient at week 4, 16, 48 and 96.

FIGURE 4.

Alluvial plot illustrates the flow of patients between different response categories on ¹⁸F-FDG PET/CT scan across four evaluation time points: at week 4 (W4), week 16 (W16), week 48 (W48) and week 96 (W96).

CMR = complete metabolic response; HGR = heterogeneous response; PMD = progressive metabolic disease; PPD = pseudoprogressive disease; SMD = stable metabolic disease; PMR = partial metabolic response; n = number of patients

Among the 71 included patients, 56 (79%) developed irAEs, including 13 (18%) with grade 3 or higher irAEs. All irAE, their number and time to onset, are presented in Table 2. One (2%) patient died of immune-related encephalitis. Due to irAEs, 7 (10%) patients were hospitalized. Three (4%) patients were diagnosed with autoimmune disease prior to ICI initiation: one had vitiligo, one had scalp psoriasis, and one was diagnosed with rheumatoid arthritis at the time of ICI initiation. None of them experienced an exacerbation of their autoimmune disease or needed special treatment for that reason.

The median OS was 25.3 months (95% CI = 17 months - NR) in patients with irAEs and 4.6 months (95% CI = 3.7 months - NR) in patients without irAEs (p = 0.004) (Figure 6A). Univariate Cox analysis showed a HR 0.9 (95% CI = 0.18–0.75; p = 0.006). The median OS was not reached (95% CI = 23.7 months - NR) in patients who experienced cutaneous irAEs and was 8.2 months (95% CI = 4.6–17.8) in patients without cutaneous irAEs (p < 0.0001) (Figure 6C); Cox analysis showed a HR 0.36 (95% CI = 0.19–0.66; p = 0.001). Using Kaplan-Meier analysis, a significant statistical difference in OS was not observed between patients with and without higher-grade irAEs (p = 0.783) (Figure 6B), endocrine irAEs (p = 0.7) (Figure 6D) or immunerelated thyroiditis (p = 0.711) (Supplementary Figure 1).

FIGURE 6.

Kaplan-Meyer curves of the median overall survival (OS) over time in patients with metastatic melanoma treated with immune checkpoint inhibitors according to (A) occurrence of immune-related adverse events, (B) occurrence of serious immune-related adverse events, (C) cutaneous immune-related side effects, (D) immune-related endocrine immune-related side effects. The blue and pink shading reflects the 95% confidence intervals for respecting groups.

FIGURE 7.

A 67-year-old male patient, diagnosed with metastatic NRAS-mutated cutaneous melanoma with lung metastases in January 2021, was treated with pembrolizumab in the first-line setting. Serial 18F-FDG PET/CT scans were obtained per the study protocol. The images above show the maximal intensity projection (MIP) on the baseline PET/CT (left), on the week 4 evaluation PET/CT (middle) and at the endpoint of the study (right). The images below show transverse sections of the lungs in different planes, revealing three FDG-avid metastatic nodules in the right lung (lower left images, red arrows), only small nodules with no FDG uptake on week 4 PET/CT (lower middle images, yellow arrows), a complete metabolic response, and no residual nodules found at the end point of the PET/CT images with persistent complete remission (lower right images).