Tim Illidge1, Elizabeth Phillips2, Nicholas Counsell2, Ruth Pettengell3, Peter Johnson4, Dominic Culligan5, Bilyana Popova2, Laura Clifton-Hadley2, Andrew McMillan6, Peter Hoskin1, Sally Barrington7, John Radford1
1The University of Manchester and the Christie NHS Foundation Trust, University of Manchester, Manchester, UK,2Cancer Research UK and UCL Cancer Trials Centre, University College London, London, UK,3Haemato-oncology Department, St George‘s, University of London, London, UK,4Cancer Research UK Centre, University of Southampton, Southampton, UK,5Haematology Department, Aberdeen Royal Infirmary, Aberdeen, UK,6Haematology Department, Nottingham City Hospitals NHS Trust, Nottingham, UK,7King's College London and Guy's and St Thomas’ PET Centre, School of Biomedical Engineering and Imaging Sciences, Kings College London, London UK
Background: Accurate stratification of patients (pts) to facilitate individualised treatment approaches is an important goal in early stage (ES) Hodgkin lymphoma (HL) where cure rates are high but late treatment related toxicity undermines long-term survival. In the RAPID study neither EORTC/GHSG baseline prognostic scores nor PET score (Deauville score 1 vs 2) predicted outcomes for the 75% of pts achieving complete metabolic remission (CMR) after ABVD (Barrington et al, submitted). Retrospective studies have shown that baseline maximum tumour dimension (MTD) predicts outcomes in ES-HL. We performed a subsidiary analysis of the RAPID trial to assess the prognostic value of baseline MTD in pts achieving CMR after chemotherapy.
Methods: 602 pts with stage IA/IIA HL and no mediastinal bulk were recruited. Pts in CMR after 3 cycles of ABVD were randomised to receive involved field radiotherapy (IFRT; n = 209) or no further therapy (NFT; n = 211). Baseline MTD was assessed by CT and reported locally. Cox regression was used to assess the association between MTD and event-free survival (EFS; progression or HL-related death).
Results: For pts in CMR randomised to NFT there was an association between MTD and EFS (HR = 1.02, 95% CI: 1.00–1.04; p = 0.04) i.e. an approximate 2% increase in risk per mm increase in MTD. The largest effect size and strongest statistical significance was seen with an MTD threshold of ≥ 50 mm, albeit with small numbers of pts/events across groups in these exploratory analyses (Table 1). A similar but non-significant effect size was seen in pts randomised to IFRT, with only 9 HL events in this group (HR = 1.02, 95% CI: 0.99–1.04; p = 0.19). Similar effects were observed for PFS but were not statistically significant with inclusion of non-HL deaths. Only 14/43 (32.6%) HL events occurred in PET positive pts (n = 145) with no evidence of an association between MTD and EFS (HR = 0.99, 95% CI: 0.96–1.01; p = 0.29).
Conclusion: This study demonstrates a clear association between MTD and EFS in pts in CMR after ABVD. A MTD threshold of 50 mm best stratified risk in RAPID pts receiving chemotherapy alone. More data is needed to confirm the relevance of MTD in pts receiving combined modality therapy. Although CT scans were not centrally reviewed, this real world assessment of MTD is directly relevant to clinical practice and informs the design of our follow-on study to RAPID with respect to the targeted use of radiotherapy in pts in CMR and with baseline MTD ≥ 50 mm.