Can Biomarkers Predict Treatment Outcomes in Hodgkin Lymphoma Where PET-Adapted Strategy Fails?

For patients with Hodgkin lymphoma, early or interim fluorodeoxyglucose (FDG)–PET scanning after two cycles of chemotherapy (PET-2) has been shown to be an effective predictor of treatment response – and a marker of when treatment can safely be escalated or de-escalated. According to a study published in The Lancet Haematology, adding biomarker information to the PET-2 scan prognostic model could improve the predictive value of PET scanning and identify patients who may have treatment failure even with a PET-2 negative scan.

Claudio Agostinelli, MD, from the Bologna University School of Medicine in Italy, and authors found that, even though “no other factor was better than a positive PET-2 scan in predicting treatment failure, the association of biomarkers with PET-2 increased the negative predictive value of PET-2 alone.”

The study included a training set of 208 patients with Hodgkin lymphoma treated with ABVD (doxorubicin, bleomycin, vinblastine, and dacarbazine) chemotherapy in whom factors predictive of treatment outcome were assessed; the results in this cohort were then validated in a fully matched independent cohort of 102 patients with Hodgkin lymphoma (validation set).

For both sets, inclusion criteria were:

  • the availability of a representative tissue sample collected at diagnosis
  • treatment with ABVD with or without radiotherapy
  • baseline staging and interim restaging after two ABVD courses with FDG–PET
  • no treatment change based solely on interim PET result
  • HIV-negative status

The investigators used Cox multivariate analysis classification and regression tree to compare the predictive values of tissue biomarkers in neoplastic and microenvironmental cells with that of PET-2, as well as to assess the biomarkers’ ability to correctly classify patients whose outcome was incorrectly predicted by PET-2.

The biomarkers analyzed included BCL2 and p53 expression in Hodgkin Reed Sternberg cells (HRSCs) and FOXP3 and PD1 expression in microenvironmental cells.

Long-term treatment outcomes were similar between the training set and the validation set. In the training set, 31 (15%) of 208 patients had disease progression and 18 (9%) of 208 patients relapsed. In the validation set, 19 (19%) of 102 patients had disease progression and three (3%) of 102 patients relapsed.

Multivariate analyses revealed that PET-2 was the only factor that maintained prognostic significance for both progression-free survival (PFS; hazard ratio [HR] = 33.3; 95% CI 13.6-83.3) and overall survival (OS; HR=31.3; 95% CI 3.7-58.9). Disease stage, FOXP3 expression, and p53 expression had significant associations only for PFS, with a better survival with FOXP3 and a worse survival with p53. See TABLE for full results of the multivariate analysis.

In the training set, no factor was better than positive PET-2 scan for predicting treatment failure, and no factor was able to correctly reclassify patients who, despite a negative PET-2 scan, ultimately had treatment failure.

Adding biomarker information to PET scan results also increased the sensitivity of a prognostic model compared with using PET-2 alone (78% vs. 59%).

These findings were reproduced in the validation set, the authors noted. They added that tissue biomarker assessment was feasible, cost effective, and reproducible, with most of the antibodies included in the panel routinely used in pathology labs.

“The algorithm correctly predicted the response to treatment in more than half of the patients who had a relapse or disease progression despite a negative PET-2 scan, thus increasing the negative predictive value of PET-2,” Dr. Agostinelli and authors reported. “This finding keeps with the preliminary results in interim PET response-adapted clinical trials in advanced classic Hodgkin lymphoma, pointing toward a non-negligible proportion of treatment failures in the interim PET-negative group treated with standard ABVD.”

The authors noted the study’s retrospective nature as a potential limitation, and wrote that the results warrant prospective validation in future studies or clinical trials, and may lead to a “more appropriate risk-adapted treatment.” Other limitations include the generalizability of results and availability and reproducibility of biomarker testing.


Agostinelli C, Gallamini A, Stracqualursi L, et al. The combined role of biomarkers and interim PET scan in prediction of treatment outcome in classical Hodgkin’s lymphoma: a retrospective, European, multicentre cohort study. Lancet Haematol. 2016;3:e467-79.

TABLE. Results of Multivariate Analysis for Progression-Free and Overall Survival
N Hazard ratio (95% CI) p Value
Progression-free survival
Ann Arbour Stage
I 4
II 70 1.0 0.01
III 31 2·1 (1·7–2·5) 0.01
IV 24 3·6 (3·2–3·9) 0.01
Low FOXP3 expression (<55/HPF) 35
High FOXP3 expression (≥55/HPF) 94 0·3 (0·1–0·7) 0.005
Low p53 expression (<25%) 113 1.0
High p53 expression (≥25%) 16 2·6 (1·0–6·6) 0.04
PET-2 negative 108 1.0
PET-2 positive 21 33·3 (13·6–83·3) <0.0001
Symptoms* 129 ··† 0.63
White blood cell count 129 ··† 0.68
International Prognostic Score 129 ··† 0.52
BCL2 129 ··† 0.23
PD1 129 ··† 0.27
Overall survival
PET-2 negative 108 1.0
PET-2 positive 21 31·3 (3·7–58·9) 0·002
Symptoms B* 129 ··† 0.99
White blood cell count 129 ··† 0.58
BCL2 129 ··† 0.23
FOXP3 129 ··† 0.070
HPF = high power field

*Symptoms were presence of night sweats, unexplained fever of >38°C, and weight loss of more than 10% of ideal weight during past 6 months.

†Variables not entered in model.