Predictive Value of ¹⁸F-Fluorodeoxyglucose Positron Emission Tomography (FDG PET) Performed after Two Cycles of Standard Chemotherapy (CT) on Treatment Outcome in Hodgkin Disease.

Objectives: Several prognostic models based on simple clinical variables have been proposed for Hodgkin’s Disease (HD); however, when applied in a prospective way, they showed unsatisfactory predictive value and scarce reproducibility. The prognostic role of early evaluation of treatment response by TC or Gallium scan has been proved in the past. We report here the preliminary results of a clinical trial on the prognostic role of FDG PET scan performed very early during treatment in advanced stage HD patients (pts), treated by conventional, standard-dose, CT.

Matherials and methods: Starting from january 2002, 55 new HD pts were consecutively enrolled into the trial; 42 completed the program and are valuable for the analysis. Pts characteristics were: mean age 34.9 years (16-79), advanced disease (stages IIB-IVB) in 24, and stage IIA in 18. Bulky and extra-nodal disease were recorded in in 12 and 11 pts, respectively. Histopathology was: NS in 36, LP in 3, MC in 2 and LD in 1. All pts were staged at baseline with TC scan, bone marrow trephine biopsy and FDG PET scan (PET-0); they were re-staged after 2 CT courses and at the end of the treatment, including radiotherapy, by TC scan and PET scan (TC-2, PET-2 and TC-8, PET-8, respectively). Standardized Uptake Value (SUV) was calculated in all FDG PET scans. PET-0 and PET-8 were considerate positive when the SUV value was ≥ 3 and PET-2 when the SUV ratio PET-2/PET 0 was > 33%. CT was ABVD in 36 pts, escalated BEACOPP in 5 and MOPP/EBV/CAD in 1. In 19/42 additional radiotherapy was given. The mean interval between the end of the 2nd CT course and PET-2 was 11.5 days (6-32); the interval between the end of the therapy (either CT or radiotherapy) and PET-8 was never shorter than 50 days.

Results: The mean follow-up from the final restaging was 328.7 days (11-690). Mean SUV value of PET-0 was 12.04. At the end of the program 38 pts were in CR and 4 in progression; one relapsed 13 months after CR entry. TC-2 showed PR in 37 pts and CR in 5. By contrast, PET-2 was positive in only 5 pts: two of them, with a mean SUV-2/SUV-0 ratio of 26.5% (20-33), showed a progressive reduction of SUV in subsequent PET scans, up to a complete negativity: both are in continuous CR. Two out of three pts with a mean SUV reduction to 61% (49-85) of the basal values progressed after the 5th and the 8th CT course, respectively; the third relapsed in CR one year after CT completion. 34/37 (92%) pts with a negative PET-2 showed a PET-8 persistently negative and remained in CR; 2 progressed during CT and one relapsed 12 months after the end of the therapy. Upon assuming that a PET-2 is positive when the ratio of SUV-2/SUV-0 is >33%, the Predictive Positive Value (PPV) of a PET-2 was 100% and the Predictive Negative Value (PNV) was 92%. The sensitivity of PET-2 was 50%, the specificity was 100% and the overall accuracy 93%.

Conclusions: The FDG PET scan performed very early during therapy predicts the treatment outcome in most pts. (39/42: 93%), with a PPV value of 100% and a PNV value of 92%. However, since most relapses of HD occur within 24 months from diagnosis, definite conclusions will be drawn after a minimum follow-up of two years.