Revisiting the predictive role of 18F‐fluorodeoxyglucose positron emission tomography/computerized tomography on treatment outcome in early‐stage favorable Hodgkin lymphoma

In the present manuscript Gallamini et al. comment the results of three large, phase III, randomized clinical trials in early‐stage favorable Hodgkin Lymphoma (HL), aimed at exploring the non‐inferiority of ABVD chemotherapy alone compared to combined‐modality treatment with ABVD and Involved Field/Node Radiotherapy (INRT). The authors also report the preliminary results of risk‐stratification in the first 60 enrolled patients in the phase 2, prospective, international study RAFTING: RAdiotherapy Free Treatment IN Good‐prognosis early‐stage HL (National Trial Identifier 04866654). In this trial patients are stratified, before treatment onset, according to the risk of therapy failure in a single patient basis, taking into account non only the results of interim and End‐of‐Therapy PET, but also the value of new metrics extracted from the baseline PET/CT such as the Large Nodal Mass (LNM) and Total Metabolic Tumor Volume (TMTV). Treatment intensity, consisting in ABVD chemotherapy, INRT and Nivolumab maintenance, is modulated on the presence/absence of the above factors, in a personalized‐medicine approach. The most frequently detected factors driving treatment intensity were LNM and TMTV, while the results of interim and end‐of‐treatment PET were also determinant, albeit in a lower percentage of cases.

radiation therapy (RT) and more commonly in patients treated with combined-modality treatment (CMT) with chemotherapy and radiotherapy compared to chemotherapy (CT) alone. 3 the millennium turnaround significant efforts have been made to mitigate the toxic effects of RT in early-stage HL (eHL) by reducing the surface of irradiated areas and the doses of RT, by the introduction of modern conformal radiotherapy and modulated intensity of radiation and adopting 3-D imaging to delineate Clinical Target Volume to be irradiated.Preliminary reports have been published showing that a reduction in TRM in HL treated with radiation, with or without chemotherapy was indeed observed, but patient follow-up in these studies was too short to draw any definite conclusion. 4erefore, a strong unmet clinical need still exists in eHL therapy to maintain a very high treatment efficacy while sparing late-onset severe morbidity and mortality related to RT.After a minimum follow-up after randomization of 60 months, the 5-Y PFS in the standard and experimental was: 94.6%; (95% CI: 91.5%-97.7%)and 90.8% (95% CI 86.9%-94.8%),respectively.However, the non-inferiority margin value fell astride the 95% CI intervals of the 5-Y PFS value of NFT arm and therefore the noninferiority of NFT treatment strategy could not be demonstrated.
The non-inferiority limit for chemotherapy alone of patients with a negative interim PET after only two ABVD cycles (PET-2) was higher (10%) in the European Organization for Radiotherapy of Cancer (EORTC)/Lymphoma Study Association/Fondazione Italiana Linfomi H10 trial.In this study, upon treatment stratification in favorable (F) and unfavorable (U) eHL according to the criteria proposed by the EORTC, enrolled patients were randomized to a PET-adapted experimental arm both in PET-2 negative (with a noninferiority design) and in PET-2 positive arm (with a superiority design) or to non-PET adapted conventional treatment arm with CMT in both strata.Briefly, PET-2 negative patients were randomly assigned to CT alone (de-escalation arm) with 4 (F arm) or 6 ABVD (U arm) cycles, or to a conventional CMT treatment with 3 or 4 ABVD courses plus IFRT, respectively. 6As mentioned, the noninferiority margin in both de-escalation arms was set to 10%.The non-inferiority arm was stopped after the pre-planned interim analysis, due to a much higher incidence of events in the chemotherapy alone compared to the standard CMT arm, indicating the futility of non-inferiority approach. 7Nevertheless, the long term disease control of PET2 negative patients treated with CT alone was relatively good with 5-y PFS of 87.1% and 89.6% in F and U group, respectively, while the outcome of patients treated with CMT was significantly superior, with a 5-Y PFS of 99.0% and 92.1%, with an Hazard Ratio (HR), 15.8; (95% CI, 3.8-66.1);and HR, 1.45; (95% CI, 0.8-2.5),respectively, favoring CMT over CT alone in both strata. 6re recently, the results of the large randomized international study Hodgkin Disease 16 conducted by the German Hodgkin Study Group (GHSG) have been published. 8 Neither of the two objectives was reached: the 5-Y PFS of patients with a negative PET-2 in both arms treated with CMT or NFT was 93.4% (95% CI 90.4%-96.5%)versus 86.1% (95% CI 81.4%-90.9%),with a delta of 7.3 (−13.0% to −1.6%) and an Hazard ratio of 1.78 (95% CI: 1.02-3.12),but the inferiority margin of 83.4% fell astride the 95% CI of the 5-Y PFS of patient treated with CT alone, and, once again, non inferiority of CT alone versus CMT could not be demonstrated.
When the threshold for a positive PET-2 was more correctly set between DS 3 and 4, the 5-Y PFS of patients treated with CMT and a negative PET-2 was 93.1% (95% CI 90.7%-95.5%)and for patients with a negative PET-2 addressed to NFT 80.9% (95% CI 72.2%-88.7%),with a difference of −12.2% (−21.3% to −3.1%).The inferiority margin was set at 88.1% and again this fell astride the 95% CI of patients with a negative PET-2 treated with CMT.However, the difference of 5-Y PFS between the two groups of PET-2 negative patients was highly significant, with a Hazard ratio of 2.94 (95% CI 1.63-5.31),p = 0.0004.
6][7][8] This suggests that RT applied "on demand" for a limited relapse might be equally effective as the pre-planned radiation in the frontline CMT modality one.

| COULD WE IMPROVE THE PREDICTIVE ROLE OF PET-2 ON TREATMENT OUTCOME IN EARLY-STAGE HL?
As well known, in early-stage HL treated with standard CMT the positive predictive value (PPV) of PET-2 is reportedly as low as 50%, probably because of the rescue effect of RT following CT in a substantial fraction of patients.As a matter of fact, in the seminal paper on the overall PV of PET-2 in HL, Hutchings et al. reported a PPV for PET after 2-3 cycles of ABVD in early-stage HL of only 30% whereas the negative predictive value (NPV) was as high as 95%. 9Similarly, in another prospective study conducted in a cohort of 88 stage I-II nonbulky HL patients treated with a nonstandard regimen (doxorubicin, vinblastine, and gemcitabine), Straus et al., upon adopting the International Harmonization Project criteria, reported a NPV and a PPV of PET-2 of 88% and 54%, respectively (P = 0.0009).Similar results were obtained upon raising the bar for a positive PET-2 in the Deauville 5PS between score 3 and 4 (85% vs. 50%).On the other hand, the PPV and NPV of End-of-Therapy PET in the same study was 73.3% and 86.1%, respectively. 10The RAdiotherapy Free Interestingly, these patients, in most cases (4/5) with a negative, or a mildly positive PET-2 (scored 4) in only one of them, showed a frankly positive End-of-therapy PET, scored 5 in most cases (See Figure 1).All the PET images reported in this preliminary report as all the PET scans performed in the RAFTING trial underwent blinded independent central review by a panel of experts.A positive EoT-PET, although not common, has been already reported in 3%-6% of PET-2 negative eHL patients after ABVD chemotherapy alone. 12,13Very interestingly, involved-site radiotherapy, at the dose of 20-40 Gy, administered as "rescue" therapy in patients with a positive PET after 3-4 cycles of ABVD was able to achieve a longterm disease control (4-Y PFS) in only 84%-54% of them.This F I G U R E 1 Patients enrolled in the RAFTING trial with a positive end-of-therapy PET after 4 ABVD cycles.§ Age is part of the so-called modified European Organization for Radiotherapy of Cancer (EORTC) criteria for early-stage Hodgkin Lymphoma (HL) prognosis, defined as follows: (1) LNM defined a s single or conglomerated multiple nodal lesions with the largest diameter ≥5 cm. ( 2 observation seems to suggest that the administration of radiotherapy alone "on demand" for good-prognosis, PET-2 negative eHL patients achieving less than CR or relapsing after ABVD chemotherapy could not be an adequate treatment to rescue all the patients and supports the addition of immunotherapy after radiotherapy in these patients.
Finally, very important, up to 51% of the patients stratified for risk had a large nodal mass in baseline PET, thus deserving a fourcourses ABVD treatment and and 19% of the patient had a high TMTV value, and were addressed to a treatment with the triplet, ABVD, INRT and Nicolumab.In conclusion: these preliminary data, in line with the published literature, support the concept that the optimal strategy for a PET-adapted strategy in eHL should be based not only on interim, but also on end-of treatment PET.Moreover, the baseline PET-derived metrics, LNM and TMTV are emerging and important risk factors related to tumor burden in eHL.

CONFLICT OF INTEREST STATEMENT
The Authors do not have conflicts of interest to declare.
Three important randomized clinical studies have been recently published in early-stage HL (eHL) favorable, aimed at exploring the feasibility and safety of a radiation-free treatment strategy based on standard chemotherapy alone with Doxorubicin, Bleomycin, Vinblastine and Dacarbazine (ABVD) in favorable eHL with a negative interim-positron emission tomography after 2 (PET-2) or three cycles of ABVD.The RAndomized Phase III trial to Determine the Role of 18F-fluorodeoxyglucose-PET Imaging in Clinical Stages IA/IIA Hodgkin's Disease (RAPID) trial was launched in 2003 with the objective to assess whether RT could be safely omitted in patients with non-bulky clinical stage IA-II A eHL and a negative interim PET after three ABVD cycles. 5After three ABVD cycles, 420 of the 426 patients with negative PET findings were randomized to involved-field radiotherapy (IFRT: 209 patients) or no further therapy (NFT: 211 patients).The non-inferiority margin was set on a Progression Free Survival at 5 years (5-Y PFS) of 87.6%.
Briefly, 1115 patients with eHL in stage I-II without risk factors according to GHSG (large mediastinal mass, extranodal lesions, Erythrocyte Sedimentation Rate (ESR) >50 mm. or >30 mm in presence of B symptoms, or ≥3 nodal regions) have been enrolled and treated with 2 ABVD courses followed by a PET-2.Patients in the standard arm of the study were then addressed to Involved-node radiotherapy (INRT), whatever the results of PET-2.In the experimental arm patients with a negative PET-2 were randomly assigned to consolidation INRT or NFT.Patients with a positive PET-2 were treated with INRT.As in the RAPID trial, the 5-point Deauville Scale (5-PS) was also adopted for PET-2 interpretation in this study, with a cutoff for a positive PET-2 set between Deauville Score (DS) 2 and 3.The trial was powered to exclude an inferiority of ≥10% in 5-year PFS of chemotherapy alone with ABVD, compared with CMT in PET-2 negative patients, in a noninferiority study design, and to detect a 5-year PFS difference of ≥5% between PET-2-positive and -negative patients receiving CMT.
Treatment IN Good-prognosis early-stage HL (RAFTING) trial (see below) has been conceived moving from the assumption that the combined results of PET-2 and post-ABVD PET (EoT-PET) provides additional information treatment outcome to improve the PPV of PET-2 alone in patients with good-prognosis early-stage HL treated with two or four cycles of ABVD chemotherapy alone.The RAFTING trial (EUDRACT n°2020-002382-33; National Clinical Trials Identifier n°04866654) is an example of a personalized medicine approach, aimed to deliver the appropriate treatment intensity to early-stage nonbulky HL patients, moving from the best risk-stratification so far available in a single-patient basis.In brief, most patients with low or intermediate low-risk (70%-75%) stage I-IIA nonbulky HL are treated with CT alone, 2 or 4 cycles of ABVD, depending on the absence or presence of at least one modified EORTC (M-EORTC) criterion, respectively.The m-EORTC criteria consist in the classic EORTC criteria in which the standard mediastinal bulky is replaced by a "Large nodal mass".The latter is defined as a single or conglomerated nodal mass with a largest diameter ≥5 cm, measured either on CT or on Maximum Intensity Projections standardized images in PET/CT.The other factors included in the standard EORTC criteria are: (1) ≥ 4 lymph nodal regions, age >50 years, ESR >50 mm Hg.After CT alone most patients are addressed to a clinical follow-up.INRT is delivered to patients failing chemotherapy (less than complete remission [CR] after ABVD or relapsing with the pattern of limited relapse), at the dose of 36 Gy, followed by Nivolumab 240 mg.Twice in a month for 24 doses.INRT is also delivered, as part of frontline therapeutic strategy, in high-risk patients, identified either by a positive interim PET after 2 ABVD cycles (PET-2) or a high Metabolic Tumor Volume measured in the baseline PET/CT.In this patient subset, the frontline therapy consists in the triplet with (1) ABVD x 4, (2) INRT at the dose of 20 or 30 Gy, depending on the absence or presence of at least one m-EORTC criterion, and (3) Nivolumab, 240 mg i.v.every 15 days for 1 year.In patients who entered CR after CT alone, cell-free tumor DNA circulating in the blood is checked regularly during the follow-up of the first 2 years with the aim of detecting, as early as possible, an impending relapse.The latter must be confirmed by a PET/CT scan.The results of the preplanned interim analysis upon enrollment of the first 40 patients was presented during the XII°International Symposium on HL held in Cologne from October 22nd to 24th 2022.Surprisingly, a small fraction (5, 8%) among the first 60 enrolled patients stratified for risk of treatment failure showed a frankly positive end-of-therapy (EoT) PET performed for restaging purposes after CT alone. 11 ) Age >50 years.(3) ≥4 different nodal regions.(4) Erythrocyte Sedimentation Rate (ESR) >50 mm.Abbreviations: DS, Deauville Score; INRT, Involved Nodal Radiotherapy; LNM, Large Nodal Mass, with a ⌀ ≥ 5 cm.; MTV, Metabolic Tumor Volume >75 mL.610 -GALLAMINI ET AL.