To the Editor:

Pegfilgrastim (Amgen Inc, Thousand Oaks, CA) is a pegylated formulation of filgrastim with a longer half-life that allows single-use per cycle. These granulocyte colony-stimulating factors (CSFs) are approved by the United States Food and Drug Administration to reduce the incidence of infection following myelosuppressive regimens that have a significant risk of febrile neutropenia. Filgrastim is also an integral part of adjuvant chemotherapy with doxorubicin and cyclophosphamide (AC) followed by paclitaxel given every 14 days (dose dense) as described in the Cancer and Leukemia Group B trial C9741.¹ Until now, the use of pegfilgrastim in the period between 14 days before and 24 hours after cytotoxic chemotherapy was not recommended due to “… the potential for an increase in sensitivity of rapidly dividing myeloid cells to cytotoxic chemotherapy.”²

Burstein et al³ recently reported in the Journal of Clinical Oncology the results of a phase II trial with the same dose-dense regimen supported by pegfilgrastim, and used historical data from trial C9741 with seven doses of filgrastim (days 3 to 10) as a comparison.¹ Pegfilgrastim was well tolerated, though a modest leukocytosis was observed after each cycle of paclitaxel and was attributed in part to the concomitant use of dexamethasone. Those investigators addressed this effect by reducing the doses of dexamethasone to just 10 mg intravenously in cycles two to four of dose-dense paclitaxel.³ In this communication, we describe three case reports from our clinical practice at The Sidney Kimmel Cancer Center at Johns Hopkins (Baltimore, MD) associated with the use of pegfilgrastim to support adjuvant chemotherapy for breast cancer. One report of hyperleukocytosis during chemotherapy is of major concern. Two other reports of late occurrence of fever and grade 4 neutropenia several weeks after the last dose of chemotherapy and pegfilgrastim have unclear clinical significance at present.

Patient 1 is a 67-year-old white woman recently treated with preoperative dose-dense AC for four cycles followed by paclitaxel for four cycles, at first supported by pegfilgrastim 6 mg subcutaneously (SQ) on day 2, and who was hospitalized with dyspnea and hyperleukocytosis 48 hours after the second cycle of paclitaxel despite a previous reduction in the dose of dexamethasone. Her total WBC count was 5,170/mm³ at baseline and she had gradual elevation of WBC in subsequent cycles of AC (cycles two to four, day 1 WBC were 7,100/mm³, 13,410/mm³, and 14,780/mm³, respectively). Her WBC on cycle five, day 1 (first dose of paclitaxel) after two doses of dexamethasone 20 mg orally as premedication was 21,360/mm³, and she received another 20 mg intravenously due to the late start of paclitaxel infusion followed by pegfilgrastim on day 2. Her WBC on cycle six, day 1 (second dose of paclitaxel) was 38,000/mm³ and dexamethasone was limited to a single dose of 20 mg intravenously followed by pegfilgrastim on day 2. She returned on day 3 complaining of sudden onset of dyspnea at rest and had a WBC of 123,310/mm³. Pulse oxymetry was 96% and a spiral computed tomography showed no pulmonary infiltrates or pulmonary embolism. She was admitted for observation, hydration, and consideration of leukopheresis,² but gradually improved and was discharged after 24 hours. Her WBC decreased to 48,580/mm³ on day 9 and to 25,550/mm³ on day 15 (cycle seven, day 1). This time, she received just 10 mg of dexamethasone orally before the third dose of paclitaxel, and pegfilgrastim was replaced by a brief 4-day course of filgrastim 300 μg SQ starting on day 3. She returned for the last dose of paclitaxel (cycle eight) with a WBC of 31,480/mm³, and this time received no CSF and no dexamethasone. She did well and her WBC on day 20 had returned to baseline values (5,600/mm³).

We also observed episodes of fever and grade 4 neutropenia in two patients 3 and 6 weeks after the last dose of adjuvant chemotherapy and pegfilgrastim, and neither one had any other sign or symptom of a viral or bacterial infection. The first episode was in a 32-year-old black woman (Patient 2) who received four cycles of TAC (AC plus docetaxel) every 21 days with pegfilgrastim 6 mg SQ on day 2. Her absolute neutrophil count (ANC) before cycle one was 4,600/mm³ and increased to 15,240/mm³ on day 1 after dexamethasone premedication. She tolerated her chemotherapy well (cycles two to four, day 1 ANC were 14,396/mm³, 7,150/mm³, and 10,750/mm³, respectively), and had an ANC of 3,230/mm³ on cycle four, day 29. She presented on day 41 with fever and ANC of 490/mm³, and was admitted for intravenous antibiotics. Cultures remained negative and she was discharged 48 hours later with an ANC of 853/mm³ to complete a course of oral antibiotics at home. ANC reached 1,850/mm³ by day 50 and remained below 2,000/mm³ until approximately day 100. She started adjuvant tamoxifen after postmastectomy radiation therapy and remains well.

The second episode was in a 32-year-old white woman (Patient 3) who received four cycles of AC every 14 days with pegfilgrastim 6 mg SQ on day 2. She had a cycle one, day 1 ANC of 3,300/mm³ and tolerated therapy quite well (cycles two to four, day 1 ANC were 4,970/mm³, 4,720/mm³, and 4,730/mm³, respectively). She presented with fever and an ANC of 200/mm³ on cycle four, day 20 and received an outpatient course of oral antibiotics. She remained neutropenic for at least 14 days (ANC day 25 of 0/mm³, day 30 of 31/mm³, day 34 of 342/mm³, and day 41 of 1,850/mm³).

The observations made in patients 2 and 3 have unclear significance at this time. Few of the reported trials of pegfilgrastim monitored neutrophil counts after the last dose of chemotherapy, and the true frequency of late grade 4 neutropenia events is unknown. Current guidelines do not recommend post-treatment blood tests in asymptomatic patients,⁴ and these episodes may have little short-term significance in the absence of mucosal barrier breakdown induced by chemotherapy.

The report made regarding Patient 1 may have greater significance. The package insert of pegfilgrastim describes that WBC > 100,000/mm³ was seen in less than 1% of 932 patients in its database of nonmyeloid malignancy trials, and “…not associated with any adverse effects.” However, it notes that “… leukapheresis should be considered in the management of symptomatic patients.”² No doubt a single dose of pegfilgrastim on day 2 is clearly more convenient and improves compliance as compared with an average of 11 doses of filgrastim,⁵ but it may turn out to be excessive for some patients. The early and gradual elevation of WBC seen in this patient suggests that the combination of steroids, pegfilgrastim, and a 14-day cycle with paclitaxel (a drug not usually associated with marked myelosuppression) may have contributed to the hyperleukocytosis. Of note, a reduction in dexamethasone to a single 20 mg dose in cycle six (paclitaxel dose 2) did not prevent this episode from happening.

Filgrastim is cleared via binding to the G-CSF receptor (a saturable mechanism) and glomerular filtration (an unsaturable mechanism),⁶ but the pegylated version pegfilgrastim has a significantly reduced renal clearance and increased circulating half-life.⁷ As the clearance of pegfilgrastim is primarily affected by the ANC and its rate of recovery, its pharmacokinetic and pharmacodynamic behavior has been described as self-regulating and patient specific.⁸ This characterization may be too simplistic.

It should not come as a surprise that the pharmacodynamic effects of pegfilgrastim on myeloid progenitor cells may exceed what its pharmacokinetic profile would predict, and result in a greater exposure to the DNA damaging effects of cytotoxics. However, the resulting apoptotic signals on these early precursors may potentially be suppressed while on CSF therapy,⁹ and only become manifest in the form of neutropenia several weeks after the last dose of pegfilgrastim. This would explain why simply withholding all CSF support midtherapy in patients with pegfilgrastim-induced leukocytosis may not be prudent and risks subsequent grade 3 to 4 neutropenia and chemotherapy delay. This happened to another patient of ours who had a day 1 WBC count of > 50,000/mm³ and had subsequent pegfilgrastim withheld, and a similar event was also described by Burstein et al.³ In these situations, a potentially safer and simpler intervention after a high day 1 ANC (eg, > 25,000/mm³) might be to simply switch from pegfilgrastim support to just a few doses of filgrastim, as five or fewer doses of filgrastim might be enough to support adjuvant chemotherapy.¹⁰ Prospective studies are warranted.

CSFs are clearly an advance for many patients, though the hyperleukocytosis event observed in Patient 1 raises concerns about the potential overutilization of pegfilgrastim in some patients. They are among the top-selling oncology drugs worldwide,¹¹and the United States Food and Drug Administration recently expanded the pegfilgrastim label to include the primary prophylaxis of moderately myelosuppressive regimens (febrile neutropenia risk < 20%).^12,13 These case reports should serve as a reminder for the need for continuing long-term surveillance in view of the rapid expansion of the use of CSFs in otherwise healthy cancer patients receiving adjuvant chemotherapy.

Author’ Disclosures of Potential Conflicts of Interest

Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.