Managing chemotherapy-induced neutropenia

The symposium was chaired by Professor (Dr) Theodor Dingermann of the Goethe University Frankfurt, Germany. He began by outlining the evolution of granulocyte-colony stimulating factor (G-CSF) therapy in the management of chemotherapy-induced neutropenia. G-CSF has a role in stimulating the proliferation and differentiation of neutrophil progenitor cells in the bone marrow, and in modulating the survival and functioning of mature neutrophils. These neutrophils are essential for innate immunity and the body’s defences against infection.

Professor Dingermann defined neutropenia as an abnormally low number of neutrophilic granulocytes in the blood. Neutrophils are liable to destruction by cytotoxic peptides and chemotherapy. Neutropenia is graded according to Absolute Neutrophil Count (ANC). The US National Cancer Institute neutropenia grading criteria are the widely accepted yardstick, and grades neutropenia (Table 1).

In those patients with severe neutropenia, the presence of a fever, usually with no other signs of infection, is suggestive of febrile neutropenia (FN). Professor Dingermann provided definitions of FN from the European Society for Medical Oncology (ESMO) and the Infectious Diseases Society of America (IDSA). The ESMO criteria are: Oral temperature 1 x >38.5°C or

2 x >38.0°C for >1 hour, with ANC <0.5×109/l. By contrast, the IDSA criteria are: fever, oral temperature ≥38.3°C or ≥38.0°C for 1 hour, with ANC  <0.5×109/l or expecting lower ANC in the next 48 hours.

Professor Dingermann identified CIN as the most serious haematological toxicity of cancer chemotherapy, with the risk of potentially life-threatening bacterial and fungal infections. As the normal inflammatory response is suppressed in these patients, the only indication of an infection may be a rising temperature ≥38.3°C.

Neutropenia has a significant impact on the delivery of planned chemotherapy, with neutropenic events including hospitalisation for FN, dose delays ≥7 days, or dose reductions ≥15% of dose intensity. The risk of developing FN is disease-dependent, with highest rates in colorectal cancer and small-cell lung cancer, and lowest occurence in non-Hodgkin’s lymphoma, or non-small-cell lung cancer. Similarly, severe neutropenia (SN) or FN are most likely to occur in small-cell lung cancer or breast cancer, and least likely to occur in ovarian cancer.

Professor Dingermann indicated that proactive use of G-CSF according to guidelines improved adherence to chemotherapy. G-CSFs are generally used in prophylaxis of SN and FN according to guidelines where the risk of FN is >20%. FN could be treated with antibiotics, antifungals or G-CSF.

Antibiotics are the mainstay of treatment of FN, with the type used dependent on the patient’s risk of infection. Antibiotics can be used until there is a sufficient ANC. There is a growing trend to use antibiotics for prophylaxis in ‘low-risk’ patients. Antifungals, on the other hand, are used in cases of prolonged FN. The use of G-CSF treatment is dependent on whether or not the patient received prophylactic G-CSF and the risk of severe outcome (based on special circumstances). The treatment options for the use of short-acting and long-acting G-CSF are outlined in Tables 2 and 3.

Professor Dingermann gave an overview of the molecular origins and modifications of the short-acting and long-acting G-CSFs, and the implications on their pharmacokinetics and pharmacodynamics, using filgrastim as a case in point. Filgrastim (rmetHuG-CSF=recombinant methionylated human G-CSF [Neupogen®]) is produced using bacterial cells, and is 175-amino acids long (molecular weight 18.8 kDa). Filgrastim is a non-glycosylated variant of the human G-CSF, which also carries the amino acid methionine on its N-terminus. Filgrastim is short-acting because it is rapidly degraded by proteases and its small size means it can be eliminated by first-pass renal clearance. In addition, it is recognised by antibodies and eliminated by white blood cells. Filgrastim’s serum elimination half-life is approximately 3.5 hours, meaning daily dosing is required.

Conversely, long-acting G-CSFs can be dosed once per cycle of chemotherapy. One method of converting short to long-acting G-CSF is the chemical attachment of a polyethylene glycol (PEG) moiety at a specific point on the protein sequence: a process known as PEGylation. PEGylation increases the hydrodynamic volume of the compound, increasing the effective molecule size by 5-to-10-times, and providing protection from immediate proteolytic degradation, first-pass renal clearance, antibody recognition and immune system elimination. PEGylation therefore significantly increases the half-life of the molecule.

Filgrastim is the base for the PEGylated variant, pegfilgrastim. In order to produce pegfilgrastim, a PEG chain of 20 kDa is covalently attached to the N-terminal amino acid of filgrastim, resulting in a molecular weight of approximately 39 kDa. Elimination of pegfilgrastim is not linear in proportion to the dose provided, but decreases with increasing doses. Pegfilgrastim appears to be eliminated primarily via neutrophil-mediated clearance, which reaches saturation at higher doses.

As a result, serum concentrations of pegfilgrastim fall rapidly as neutrophil count recovers.

An alternative modification of filgrastim is GlycoPEGylation, a combination of glycosylation and PEGylation. This is achieved by enzymatic synthesis in a two-step process shown in Figure 1:

Natural O-glycosylation site threonine (T) accepts N-acetyl-galactosamine (GalNAc), catabolised by GalNAc transferase.

The GalNAc residue serves as a direct acceptor for PEGylated sialic acid, forming a biologically active, chemically homogeneous glycoPEGylated protein, lipegfilgrastim, with an extended plasma half-life.

Studies comparing the pharmacokinetics of lipegfilgrastim and pegfilgrastim, in healthy volunteers, showed a higher bioavailability of lipegfilgrastim, compared with the equivalent dose of pegfilgrastim (approximately 60% bigger area under the curve). In terms of pharmacodynamics, lipegfilgrastim induced approximately 30% higher ANC response compared with the equivalent dose of pegfilgrastim.

Professor Dingermann concluded that G-CSFs play an essential role in supportive care during cancer chemotherapy by stimulating neutrophil production, and that this effect significantly reduces the incidence of FN. He advocated the advantages of using the long-acting G-CSFs, for their once-per-cycle dosing, and greater ANC response.

Professor (Dr) Rupert Bartsch of the Medical University of Vienna, Austria, presented the medical rationale for using long-acting G-CSFs. He opined that G-CSF support reduces FN risk by >94% in patients receiving chemotherapy regimens with FN risk ≥20%, and that early use of G-CSF support reduces neutropenia-associated hospitalisations by 80%.

Analysing the effect on hospitalisation and use of antibiotics, Professor Bartsch presented a meta-analysis of eight randomised control trials in breast cancer, involving 2156 patients, which showed that G-CSF support significantly reduced the risk of FN, early mortality risk, risk of hospitalisation and use of antibiotics.

In comparing long-acting and short-acting G-CSFs, he reviewed the GeparTrio study of 2439 patients, randomised to receive primary prophylaxis of ciprofloxacin, daily G-CSF, pegfilgrastim or pegfilgrastim + ciprofloxacin. The incidence of FN in all cycles was approximately the same in the ciprofloxacin and daily G-CSF groups. However, there was a 60% fall in the incidence of FN in the pegfilgrastim group, and a 70% fall in the pegfilgrastim + ciprofloxacin group, when compared with the daily G-CSF group. Both reductions were significant. The effect of pegfilgrastim was more pronounced in the first cycle, where the reductions in the incidence of FN in the pegfilgrastim and pegfilgrastim + ciprofloxacin groups were 80% and 100% respectively, compared with the daily G-CSF group.

A comparison of lipegfilgrastim and pegfilgrastim was presented in a prospective, randomised Phase III study with the primary endpoint of duration of SN in the first cycle of chemotherapy (<0.5×109/l; non-inferiority design), and secondary endpoints of incidence of FN; time to ANC recovery; incidence of hospitalisations; use of intravenous antibiotics; and quality of life. There was no significant difference the drugs in terms of the primary endpoint. However, in terms of the secondary endpoints, the mean time to ANC recovery at cycle 1 was significantly shorter in the lipegfilgrastim group than in the pegfilgrastim group.

Reviewing the role of G-CSF support in breast cancer and lymphoma, Professor Bartsch highlighted the growing incidence of cancer in the elderly. In developed countries, 40% of newly diagnosed breast cancer patients were ≥65 years of age (worldwide 35%), and that the incidence of many malignancies including breast cancer and lymphoma increases with age. However, as the elderly population has historically been under-represented in clinical trials, there are concerns that the ‘optimal’ dosing regimens identified in these studies may be intolerable for the elderly.

Older patients often have comorbidities, and the physiological processes associated with ageing change their responses to treatment, often with the resultant potential for increased side-effects. In the studies presented, long-acting G-CSF support prevented FN to the largest extent in elderly patients (≥65 years of age) undergoing chemotherapy for breast cancer. Similar benefits have also been demonstrated for elderly patients undergoing chemotherapy for aggressive forms of lymphoma (including diffuse large B-cell lymphoma and T-cell lymphoma), when they have received primary or secondary prophylaxis with long-acting G-CSF.

An overview of the impact of G-CSFs on the chemotherapeutic landscape was presented by Professor (Dr) Achim Rody, a breast cancer specialist at the University Medical Centre, Lubeck, Germany. He stated that there were 1.7 million new cases of breast cancer diagnosed worldwide each year, with over 520,000 deaths from the condition annually. He outlined the multimodal approach to the treatment of breast cancer, including biological therapy, chemotherapy, endocrine therapy, radiotherapy and surgery, and indicated that molecular sub-typing can help to identify optimal therapy options.

Professor Rody also presented the ‘vicious circle’ of cancer chemotherapy, suggesting that neutropenia was a major cause of reduced survival (morbidity and mortality), with the mortality rate of FN with hospitalisation being 9.5% (Figure 2)

He subsequently reviewed the European Organisation for Research and Treatment of Cancer (EORTC) 2010 guidelines on neutropenia and FN, reiterating the recommendation to use G-CSF as prophylaxis against chemotherapy-induced FN (Figure 3).

Table 4 shows the ESMO recommendations for G-CSF for primary prophylaxis.

Table 5 shows the special situations for the use of G-CSFs for standard therapy.

The American Society for Clinical Oncology (ASCO) and the National Comprehensive Cancer Network (NCCN) also identify reduction in FN as an important clinical outcome that justifies the use of G-CSF when the risk of FN is approximately 20% or above. They recommend primary prophylaxis for the prevention of FN in patients who are at high risk based on age, medical history, disease characteristics, and myelotoxicity of the chemotherapy regimen. Professor Rody summarised the recommendations, as shown in Table 6.

Following on from the presentations, a very lively panel discussion with audience participation ensued.

Professor Dingermann commented on the fact that in Professor Rody’s presentation, breast cancer G-CSF guidelines appeared to be adhered to much more strongly than lung cancer G-CSF guidelines. However, even in the breast cancer setting, intermediate-risk G-CSF guidelines were not strongly adhered to. Professor Rody responded that many clinicians initiate chemotherapy alone, and only introduce G-CSF once FN has developed. There is a clear need to provide education around when to use G-CSF.

Professor Dingermann closed the meeting with the overall conclusion that G-CSF, in all its forms (short-acting, long-acting, biosimilars) saves lives, and is a critical, supportive rather than curative, drug for CIN.