How filgrastim has helped control neutropenia and prevent sepsis

Before the first granulocyte colony-stimulating factor (G-CSF) filgrastim (Neupogen) was launched in the UK in 1991, little could be done to prevent the neutropenia and potentially life-threatening infection associated with some of the chemotherapy regimens of the day.

Barry Hancock, emeritus professor of oncology at the University of Sheffield, recalls that chemotherapy in the 1980s was not as intensive as many of today’s regimens but, even so, neutropenic sepsis was an expected side effect for many patients.

“Decisions about chemotherapy were tempered by the need to give treatment that was less likely to cause neutropenia, or in lower doses that reduced the risk. We hoped that if patients did become neutropenic it wouldn’t be a problem and we warned them to seek urgent help from their cancer specialist if they had any sign of fever,” he says.

“If patients did get neutropenic sepsis we had to delay their chemotherapy and, without G-CSF, it took longer for their white cell count to return to an acceptable level for their next cycle,” he adds.

Discovery of G-CSF

Colony-stimulating factors (CSFs) were first discovered in the 1960s when researchers in Australia and Israel independently found that in vitro growth of bone marrow and blood cells was stimulated by unidentified factors diffusing from leukaemic, embryonic and other cells with which they were cultured.¹ It took another 20 years to identify and purify the CSFs for in vivo studies because of the small amounts of colony-stimulating activity present in tissues and fluids that were tested.

During that time it became clear that there were four types of CSF capable of stimulating different colonies of cells: granulocyte and macrophage CSF (GM-CSF), macrophage CSF (M-CSF), G-CSF, and multi-CSF (interleukin [IL-3]).²

Mouse and then human G-CSFs were purified and advances in genetic engineering in the early 1980s led to cloning of the G-CSF gene, and expression in Escherichia coli, at the pioneering biotechnology company, Amgen, in collaboration with the Memorial Sloan-Kettering Cancer Center in New York.

In early animal studies, recombinant G-CSF was shown to boost white cell counts (mainly neutrophils) in non-human primates, and these levels were maintained throughout a four-week course of treatment.² In 1987, phase 1 clinical studies started in patients with transitional cell carcinoma of the urothelium who were undergoing chemotherapy with methotrexate, vinblastine, doxorubicin and cisplatin.² G-CSF reduced the number of days of neutropenia and need for antibiotics and significantly increased the proportion of patients who were able to complete their planned chemotherapy. Another phase 1 and 2 study of G-CSF in small cell lung cancer yielded similar, promising results.²

These were confirmed in two large placebo controlled phase 3 trials in the US and Europe in patients receiving chemotherapy for small cell lung cancer.^3,4 In patients treated with G-CSF from days 4–17 of each 21-day cycle, episodes of fever with neutropenia were halved, and infections, use of parenteral antibiotics and hospital admissions were significantly reduced.^3,4 In G-CSF-treated patients more cycles were completed without dose reduction and with fewer delays.⁴

Filgrastim enters clinical practice

The introduction of filgrastim into clinical practice was met with considerable enthusiasm, explains Professor Hancock, and it became clear that, although filgrastim did not stop the neutrophil count reaching its nadir, patients had shorter periods of neutropenia and so were less likely to become septic. “There were essentially three indications for treatment: secondary prophylaxis for patients who had already had an episode of neutropenic sepsis, treatment of neutropenia to resurrect neutrophils, especially in patients suffering prolonged neutropenia, and primary prophylaxis for patients who had accelerated or more intensive chemotherapy,” says Professor Hancock.

Quite early in the development of G-CSF, its value was recognised in restoring neutrophil levels following bone marrow transplantation and in mobilising stem cells for collection before high-dose chemotherapy.² Before long, mobilisation of peripheral stem cells with G-CSF was shown to be safe enough to be used in healthy donors, and haematopoietic stem cell transplantation largely superseded bone marrow grafts for patients with leukaemia and other types of cancer.²

Use of G-CSF in leukaemias was initially controversial because of concerns that the growth factor might stimulate cancer cells.² However, after a series of small leukaemia studies showed that G-CSF reduced neutropenia, infections and need for antibiotics without adversely affecting remission, treatment became more widely used in haematological malignancies.¹

Prophylaxis with G-CSF

Despite growing use of filgrastim and lenograstim (introduced in 1993), deaths from neutropenic sepsis in England and Wales almost doubled between 2001 and 2010.⁵ Oncologists have linked this rise to increased use of chemotherapy, more intense regimens and greater use in patients who would previously have been considered not well enough for treatment.⁵ Dose-dense chemotherapy has been shown to result in improved survival over standard chemotherapy, especially in lymphoma.⁶ But there is no evidence that primary prophylaxis with G-CSF reduces short-term all-cause mortality compared with no prophylactic G-CSF, and there is too little head-to-head data to conclude whether primary prophylaxis with G-CSF is as good or better than prophylactic antibiotics.⁵ There is some evidence that primary prophylaxis with G-CSF and antibiotics reduces infectious mortality.⁵

Professor Hancock points out that European guidelines recommend that G-CSF should only be used for primary prophylaxis in high-risk patients.⁶ This means those being treated with a chemotherapy regimen that carries a >20 per cent risk of febrile neutropenia, elderly and frail patients at increased risk, and patients undergoing high-intensity chemotherapy schedules or where delays in treatment are known adversely to affect prognosis.

Similarly, the National Institute for Health and Care Excellence guidance on neutropenic sepsis, produced in 2012 by Professor Hancock and colleagues, recommends that adults with acute leukaemias, stem cell transplants or solid tumours who are likely to become neutropenic as a result of their chemotherapy should not be routinely offered G-CSF for prevention of neutropenic sepsis unless they are receiving G-CSF as an integral part of the chemotherapy regimen or in order to maintain dose intensity.⁵ The guidelines recommend that most patients should be offered prophylaxis with a fluoroquinolone during the expected period of neutropenia rather than G-CSF, because prophylactic quinolones have been shown to reduce short-term all-cause mortality.⁵

“At the time that G-CSF was introduced, the use of prophylactic antibiotics was controversial and many people didn’t believe they prevented neutropenic sepsis. But that changed after it was shown that prophylactic antibiotics could reduce fever, probable infection and hospital admission in patients receiving chemotherapy for solid tumours or lymphoma,” says Professor Hancock.⁷

There has been considerable variation in G-CSF use in clinical practice, and Professor Hancock believes, some overuse.

“G-CSFs are wonderful drugs and they certainly work whenever they are given. But the cost-benefit is quite marginal and, even among conservative users, they have probably been overused,” he says.

The NICE guidelines are due for update in 2015, so it may be possible to see whether the 2012 guidance has impacted on use.

Long-acting formulations

Pegylated filgrastim was introduced in 2002, and has reduced treatment frequency from daily dosing during the neutropenic phase of each chemotherapy cycle to a single injection per cycle. A meta-analysis, which included five studies comparing pegfilgrastim and filgrastim, showed that the incidence of febrile neutropenia was significantly lower for pegfilgrastim than filgrastim, with a relative risk of 0.66 (95 per cent confidence interval 0.44–0.98).⁸ The authors pointed out that, although there was substantial heterogeneity of cancer type and chemotherapy regimen between studies, the range of populations was probably typical of that seen in clinical practice.⁸

“The long-acting versions of G-CSFs are also more convenient and comfortable for patients. There is only one injection per cycle and, though there is still a significant proportion who experience bone and muscle pain, it seems to be less intense with the pegylated formulations,” says Professor Hancock.

“Filgrastim and biosimilars have become part of the staple diet of chemotherapy with a definite role that is likely to continue,” he concludes. “They have helped make chemotherapy safer and easier to administer, and enabled us to increase the dose and intensity of some of our regimens. The dilemma remains over whether a short course of antibiotics is just as good and more cost-effective in many situations of prophylaxis. If you’re giving G-CSF as part of an established treatment to enable chemotherapy to be completed on time, in the right dose, that’s fine. But people who are giving it to prevent infection in patients being given standard chemotherapy with risk of significant neutropenia should think again, since antibiotics are probably the better option.”