Interferon-alpha2 (IFN) has been used in the treatment of the Philadelphia-negative myeloproliferative neoplasms (MPNs) for about 30 years and several studies have convincingly demonstrated that this agent is safe and highly efficacious in normalizing elevated cell counts. Indeed, prolonged treatment (about 5 years) may be followed by normalization of the bone marrow and low-burden JAK2V617F in a subset of patients, even being sustained for 2–3 years after discontinuation of IFN. Early treatment to reduce or eradicate the malignant clone is of paramount importance for achievement of minimal residual disease (MRD) or potentially cure in all cancers. However, in MPNs a “watch and wait” strategy is used in “low-risk” patients allowing the malignant clone to expand with the inherent risk of increasing genomic instability, sub-clone formation, resistance to treatment and disease progression. However, the “watch and wait” strategy has been challenged by reports demonstrating the potential of IFN to induce MRD in an increasing number of patients. Furthermore, these studies also indicate that early treatment with IFN increases the chance of sustained hematological and molecular remissions. However, the evidence for this statement is lacking.
By data-driven mathematical analysis of the JAK2V617F allele burden during IFN treatment of MPNs to deliver the proof of concept that supports early intervention with IFN to obtain the best possible treatment response.
We present a thorough data driven mathematical analysis of unique serial measurements of the JAK2V617F allele burden in 66 patients most being treated with IFN in the DALIAH trial and in 7 untreated patients.
Without IFN treatment, the JAK2V617F allele burden increased exponentially with a period of doubling of 1.4 year. During treatment with IFN, the JAK2V617F allele burden decreased mono- or bi-exponentially for 33 responders. Bi-exponential description improved the fits in 19 cases being associated with late JAK2V617F responses. The decay of the JAK2V617F allele burden during IFN treatment was estimated to have half-lives of 1.6 year for the mono-exponential response and in the long term 1.0 year for the bi-exponential response.
In recent years, the interest of using IFN in the treatment of MPNs has been based upon studies reporting long-term treatment with IFN to be associated with MRD in a subset of patients as defined by sustained complete hematological remissions in concert with induction of low-burden JAK2V617F and normalization of the bone marrow. The present study delivers novel information regarding the JAK2V617F kinetics during IFN-treatment. In untreated patients, the JAK2V617F allele burden increased exponentially with doubling time of 1.4 years. During IFN treatment, the JAK2V617F development followed either a mono-exponential or a bi-exponential decay for a significant proportion of patients. Analysis of the population-level responses suggests that treatment-schemes should extend longer than one year, as the efficacy of IFN treatment on the JAK2V617F allele burden cannot be determined after just one year.
In conclusion, using data-driven analysis of the JAK2V617F kinetics, we provide evidence for tumor burden reduction through early intervention with IFN, thereby challenging the “watch and wait” strategy commonly applied on low-risk MPN patients.