Effect of steroid treatment on the diagnostic yield of baseline 18f-fluorodeoxyglucose positron emission tomography in aggressive B cell lymphoma

Study design, setting and participants

The electronic healthcare database of the hemato-oncology unit of a tertiary cancer center was retrospectively reviewed for all consecutive adult patients (age ≥ 18 years) with newly diagnosed aggressive non-Hodgkin's B cell lymphoma between January 2017 and May 2020. We included patients with diffuse large B cell lymphoma, high-grade B cell lymphoma, primary mediastinal (thymic) large B cell lymphoma, and Burkitt lymphoma, all of which are associated with symptoms and a risk of organ damage and therefore often warrant prompt steroid treatment. We included patients who underwent PET/CT examination as part of the routine clinical workup. Patients were divided into those who received steroid treatment (any dose, any route) within 30 days before the PET/CT scan, and those who did not.

The study was approved by the Institutional Review Board of Rabin Medical Center.

Data collection

Clinical data were collected from the medical records, as follows: demographics, histologic diagnosis according the WHO 2016, lymphoma characteristics, steroid treatment (if any) and dose, glucose level up to 24 h prior to PET CT in milligram (mg)/deciliter (dL) and date of performance of PET/CT. The lymphoma characteristics included lactate dehydrogenase (LDH) levels (upper limit of normal in our institution, 480 U/L), ki67 proliferation index, disease stage (according to the Lugano classification), and international prognostic index (IPI) [12]. The different steroid formulations (dexamethasone, hydrocortisone etc.) were converted to prednisone equivalent doses using a standard conversion equation [13, 14].

PET/CT scan acquisition

PET CT scans were performed in various PET CT facilities in Israel (mainly in our institution), and all PET CTs were reviewed in our institution by physicians who specialize in nuclear medicine.

SUV is a semiquantitative index calculated by the ratio of FDG concentration in a selected region of interest to the injected dose which is normalized to body weight. SUVmax reflects the maximal value within a selected volume of interest; SUVmean reflects the average value in the volume of interest.

Metabolic tumor volume (MTV) is a measure of the volume of the metabolically active areas of the tumor in the volume of interest, calculated by threshold from the SUVmax. Total lesion glycolysis (TLG) is the product of SUVmean and MTV.

Segmentation was performed using CARESTREAM-PACS software, version 12.1.5.7014, which automatically defines the contour of the PET-based lesion. The cut-off is 42% of the tumoral SUVmax.

Total metabolic burden was derived from the TLG (SUVmean* MTV), with the volume of interest of the whole scan and reduction of physiological FDG uptake in the brain, heart, kidneys and bladder. The tumor volume was defined as the MTV calculated in this volume of interest.

We also calculated the SUVmax and SUVmean of the liver (3 cm diameter spherical volume of interest, segment 7) and mediastinum (1.5 cm diameter spherical volume of interest, placed on the aortic arch) and the lesion-to-liver SUVmax ratio.

Study procedure

Patients who received steroids were compared to steroid-naïve patients for clinical and PET/CT parameters. Further sub-analyses were conducted by the duration of steroid treatment (0 days, 1–3 days, 4–7 days, 8 days or more), average prednisone daily dose (0 mg, 1–20 mg, 21–59 mg, 60 mg or more), prednisone cumulative dose in the week prior to the PET/CT scan (0 mg, 1–140 mg, 141–479 mg, 480 mg or more) and prednisone mg per kilogram (kg) (< 0.5 mg/kg, 0.5 mg–0.99 mg/kg and ≥ 1 mg/kg).

Primary outcome measure

The primary outcome measure was the effect of steroid intake on SUVmax in lesions of aggressive B cell lymphoma. The effect was examined across different steroid doses and regimens.

Secondary outcome measures

The secondary outcome measures were the effect of steroid treatment on additional PET/CT parameters, including SUVmean, MTV, and tumor burden, in addition to SUVmax and SUVmean of the liver and mediastinum.

In addition, we examined whether disease and tumor characteristics influenced steroid administration prior to the PET/CT scan.

Statistical analysis

The statistical analysis was generated using SAS, version 9.4. For the two groups (steroid-naïve and steroid-treated), continuous normally distributed parameters were analyzed using t-test, and continuous non-normally distributed parameters using Wilcoxon test.

When comparing multiple groups of unpaired samples, we used the Friedman (Ranked Anova) with Tukey–Kramer correction for multiple corrections in order to test for differences between groups.

Continuous parameters are presented as mean and standard deviation (SD) or median and interquartile ratio (IQR), as appropriate. Fisher’s exact test was used to compare the value of categorical variables between study groups. Categorical variables are presented as number and percentage. Two-sided P values less than 0.05 were considered statistically significant.

留言 (0)

沒有登入
gif