Prognostic impact of pretreatment cell-free DNA concentration in newly diagnosed peripheral T-cell lymphomas

Peripheral T-cell lymphomas (PTCLs) have a poor prognosis and, to date, there are no reliable predictive biomarkers of response. In this work we explored the prognostic impact of cell-free DNA (cfDNA) concentration in 75 newly diagnosed patients enrolled in a prospective multicenter study. Pre-treatment cfDNA was strongly associated with clinical risk factors and was identified as a superior predictor for shorter progression-free survival in multivariable analysis, outweighing canonical risk parameters. Furthermore, we identified a cfDNA value above which survival worsens. In conclusion, pre-treatment cfDNA concentration represents an easily usable predictive biomarker that is highly associated with survival of PTCL patients.

G-CSF administration following peripheral blood progenitor cell (PBPC) autograft in lymphoid malignancies: evidence for clinical benefits and reduction of treatment costs

Clinical value and costs of G-CSF administration following autograft with mobilized peripheral blood progenitor cells (PBPC) were evaluated in two sequential groups of 20 patients each, treated for lymphoid neoplasms in the period February 1993 to January 1996. One group was given G-CSF (Filgrastim) (5 microg/kg/day), starting on day +1 until ANC was > 500/microl, the other received no G-CSF. All patients were conditioned with mitoxantrone 60 mg/m2 + L-PAM 180 mg/m2 and received large numbers of PBPC (median of 12 and 13 x 10(6) CD34+/kg, respectively). The median time to ANC > 500/microl was 10 days in the G-CSF group vs 14 days in controls (P < 0.0001). G-CSF was associated with a slightly faster platelet recovery (11 vs 13 days to plts > 20000/microl, P = 0.09). Median duration of fever (2.5 vs 5 days, P = 0.028), nonprophylactic antibiotics (8 vs 11 days, P = 0.019), and post-transplant hospitalization (13 vs 16 days, P = 0.0028) were also significantly reduced. The average cost per treatment in the G-CSF group amounted to about US$18241 as compared to US$21868 in the control group, implying a cost reduction of approximately 16%. Thus, G-CSF reduced morbidity with cost containment, supporting its use even if autograft is performed with large quantities of PBPC.

A single step density gradient separation for large scale enrichment of mobilized peripheral blood progenitor cells collected for autotransplantation

Peripheral blood leukocytes are becoming the preferred source of hematopoietic progenitor/stem cells for autologous transplantation. However, in vitro purging procedures are complex and expensive when applied to peripheral blood progenitor cells harvests. This is mainly due to the large quantities of nucleated cells present in leukapheresis collections. Aiming to reduce total cellularity without significant loss of CD34+ cells, we developed an in vitro cell separation procedure based on ficoll/metrizoate gradient used at a final density of 1.067 g/ml. To obtain this density, standard Lympho-prep (1.077 g/ml) was diluted with normal saline solution (NaCl 9 g/l). Twenty-six leukapheresis collections (median cellularity 21.1 x 10(9), range 2.8-60) from 14 patients with non-Hodgkin's lymphoma, multiple myeloma or plasma cell leukemia were processed (median two leukaphereses per patient). Mean (+/- s.d.) recovery of total nucleated cells, CD34+ cells and CFU-GM was 20.9 +/- 10%, 74.7 +/- 22% and 70.5 +/- 19%, respectively. Cumulative per patient progenitor cell recovery was always above 50%, and as high as 80% in 10/14 patients, while total cellularity was reduced to a median 21.5% (10-33%) of pre-separation values. Contaminating neoplastic cells, identified by immunofluorescence in five collections, were reduced by 1-2 logs. The results indicate that our density gradient separation is an effective method to reduce total cellularity prior to immunological purging, without significant loss of progenitor cells.

Peripheral blood progenitor cell mobilization in patients with primary refractory lymphoma or at first relapse: comparison with patients at diagnosis and impact on clinical outcome

Peripheral blood progenitor cell (PBPC) mobilization was evaluated in 53 patients receiving the high-dose sequential (HDS) regimen: 27 had non-Hodgkin's lymphoma or Hodgkin's disease, primary refractory or at first relapse, 26 had non-Hodgkin's lymphoma at diagnosis. Mobilization was assessed following either 7 g/m2 cyclophosphamide (48 patients) or 2 g/m2 etoposide, both followed by G-CSF (filgrastim) at 5 microg/kg/d. PBPC mobilization was significantly higher in patients at diagnosis compared to refractory/relapsed patients (median peak values of circulating CFU-GM: 25,209/ml v 4270/ml, P < 0.0001 and CD34+ cells: 286/microl v 47/microl, P < 0.0001). All patients receiving HDS as up-front treatment mobilized enough PBPC for an autograft, often requiring a single leukapheresis; whereas only 15 patients under salvage treatment with HDS were able to complete PBPC autograft. Bone marrow (BM) cells, alone or with PBPC, were needed in six patients, and autograft could not be performed in six patients. Among refractory/relapsed patients, those having a high PBPC mobilization experienced a significantly longer EFS compared to those who had not; autograft completion also significantly enhanced EFS. Thus, the use of an effective mobilizing protocol does not ensure adequate PBPC mobilization in moderately pretreated patients; low mobilization must be considered as an early sign of poor outcome in patients receiving a high-dose salvage programme.

Both early and committed haemopoietic progenitors are more frequent in peripheral blood than in bone marrow during mobilization induced by high-dose chemotherapy + G-CSF

Haemopoietic growth factor administration following high-dose chemotherapy markedly amplifies progenitor cell pool in the peripheral blood (PB). Collection and reinfusion of these cells enable rapid haemopoietic reconstitution following autograft. Less is known on engraftment potentiality of bone marrow (BM) cells taken under analogous conditions. To investigate this tissue, PB and BM were evaluated simultaneously during maximal mobilization in a series of 14 patients undergoing the HDS chemotherapy programme. A significantly higher growth of committed progenitors was found from PB rather than from BM (663 +/- 123 v 267 +/- 40 CFU-GM/10(5) MNC, respectively). Also, significantly more CFU-GM could be collected by a median of three leukaphereses, compared to those harvested from BM (158 +/- 31 v 16 +/- 4 x 10(4) CFU-GM/kg, respectively). Most mobilized CFU-GM were phenotypically immature (CD15-); in addition, circulating cells included primitive progenitors, as assessed by LTC-IC assay, or by evaluation of non-proliferating pre-CFU-GM, selected by an anti-CD71 immunotoxin. The amount of pre-CFU-GM determined by both techniques was consistently higher in PB than in BM. Moreover, a direct correlation could be established between circulating CFU-GM and primitive precursors. Thus, during optimally induced mobilization, PB contains many more haemopoietic progenitors, of both committed and primitive stages, than does BM. Under such conditions, PB is probably the best source of material for graft purposes.

Haematological support of high-dose sequential chemotherapy: clinical evidence for reduction of toxicity and high response rates in poor risk lymphomas

The toxicity and feasibility of a high-dose sequential (HDS) chemotherapy programme delivered with growth factor support were evaluated in patients with intermediate and high-grade non-Hodgkin's lymphoma (NHL) or with progressive Hodgkin's disease. The scheme includes the sequential administration of single cytotoxic drugs at very high doses followed by intensified treatment with circulating progenitor autograft. In some instances, the original HDS scheme, initially designed at the Milan Cancer Center, was partially modified and intensified with a preliminary debulking phase. The use of G-CSF (filgrastim) made toxicity in the high-dose phase acceptable and allowed good harvests of peripheral blood progenitor cells (PBPC); the use of PBPC in the final autografting phase resulted in low haematological toxicity. Of 71 patients with NHL treated at our institution with either the original or the intensified HDS version, the overall toxicity-related mortality was 5.6%, thus comparable to lethal toxicity commonly associated with conventional chemotherapy. Adequate PBPC harvests are crucial for good tolerability of the programme. Optimal harvests are generally obtained in patients without neoplastic marrow infiltration while patients with marrow disease often have a poorer mobilisation. However, an optimally time-spaced chemotherapy debulking might also restore sufficient mobilisation in these latter patients. In terms of therapeutic efficacy, HDS had produced promising results since the initial experience in relapsed patients. More recently, HDS was evaluated as first-line treatment in a series of 22 consecutive patients, presenting with advanced-stage, intermediate-grade NHL other than diffuse large cell subtype. A CR rate of 82% was obtained following HDS, with a projected survival of 86% at five years. Thus, delivery of an intensive high-dose chemotherapy programme with haematopoietic growth factor support was found to be feasible and reasonably safe. The high anti-tumour efficacy of such a scheme makes it suitable for wider applicability in all those chemosensitive tumours where a dose increase might enhance the chance of cure.