Empiric Antifungal Therapy for the Neutropenic Patient

Introduction

Since the inception of antineoplastic therapies, management of chemotherapy-induced neutropenia has posed a challenge to oncologists. At one time, neutropenic infection led to death in 25% of all cases. The introduction of safe and effective broad-spectrum antibiotics and the validation of empiric antibacterial therapy for neutropenic fever have sharply reduced infectious morbidity and mortality. However, the major challenge facing oncologists today is invasive fungal infection. Difficult to diagnose and deadly when missed, invasive fungal infection is the major infectious cause of death associated with chemotherapy-induced myelosuppression.

Invasive Fungal Infections and Chemotherapy-Induced Neutropenia

The occurrence of invasive fungal infection can be viewed as a consequence of medical progress. Prior to the advent of antineoplastic therapy, invasive fungal infection was rarely recognized in patients with neoplastic diseases.[1] The first reports of invasive fungal infection in cancer therapy noted that only after the introduction of antineoplastic therapy were these infections found in autopsied leukemia patients. In those initial reports, leukopenia, mucosal injury, and concomitant bacterial infection were associated with invasive fungal infection, and it was speculated that these were contributing factors.

With the expanding use of antineoplastic therapy, invasive fungal infection was seen more frequently in leukemic therapy as well as in the therapy of lymphomas and solid tumors. As new strategies evolved to effectively control serious bacterial infections during neutropenia, substantially more patients survived to develop invasive fungal infection. Numerous reports documented the rising frequency of invasive fungal infection during the 1960s and 1970s.[2-4] The risk of invasive fungal infection was found to be associated with the type of underlying disease—higher rates were reported in patients undergoing treatment for acute leukemia, intermediate rates in patients receiving therapy for lymphoma, and the lowest rates in patients treated for solid tumors.[4-6]

Representative frequencies of invasive fungal infection for various disease treatment categories are shown in Table 1. Data from studies conducted in the 1960s and 1970s may not accurately reflect the current risk of developing these infections because antineoplastic treatment regimens and supportive care measures have changed dramatically. Nevertheless, the data provide some perspective on the relative risks for invasive fungal infection in the absence of measures to reduce these risks (in wide use today). The data give a qualitative sense of the risk to a patient with a given disease or treatment category, and, although rates are frequently quoted by disease, it is the antineoplastic treatment strategy used in a given disease category that seems to have the greatest impact on risk of invasive fungal infection.

During the 1980s, with the introduction of new antineoplastic regimens and many of the supportive care measures in use today, the rates of nosocomial fungal infections continued to rise in multiple patient populations, including oncology patients.[7-9] In fact, the rate of nosocomial fungal infections doubled in all hospitalized patients.[7]

Candida assumed the rank of the fourth most common bloodstream pathogen, forging ahead of a number of historically important bacterial pathogens. Some reports noted that many leukemia and bone marrow transplant patients were found to have invasive fungal infection at autopsy, with an alarming proportion not being documented antemortem.[10-12] Concerns grew that invasive fungal infections were frequently underdiagnosed and that rates were substantially higher than initially thought.

Major Fungal Pathogens

Fungi are ubiquitous in nature. Although there are more than 200,000 fungal species, fewer than 150 are known to be human pathogens. Fungi are generally grouped by morphology as either yeasts or filamentous moulds.

Yeasts

Yeasts are the most frequent fungal colonizers and causes of invasive fungal infections in humans. Relatively few yeast species cause disease. The genus Candida is the most common, accounting for 60% to 70% of invasive fungal infections in cancer patients. There are approximately 200 species in the genus Candida. These organisms are present on many plants, and several species are also part of the normal flora within human and animal gastrointestinal tracts. Most normal healthy individuals carry one or more Candida species, but they are only a minor constituent of the gut flora (which is predominantly bacterial).

Historically, C albicans was the most important pathogenic Candida species. In recent years, however, non-albicans species represent an increasing proportion of Candida blood isolates in oncology as well as general hospital patient populations.[13-15] In cancer patients, approximately half of candidal fungemias today are non-albicans species, including C tropicalis, C glabrata, C parapsilosis, C krusei, and several others (Table 2).

There is a propensity for certain Candida species to cause infection in specific disease groups. C glabrata and C parapsilosis are reported in a relatively greater proportion of patients with solid tumors and lymphoma, whereas C tropicalis is encountered more frequently in patients treated for leukemia and C krusei is more common in bone marrow transplant recipients.[13]

There is also an association between certain medical practices and specific Candida species. For example, antileukemic regimens that give rise to greater mucosal barrier damage have been associated with C tropicalis. The use of polyene antifungal agents is associated with C lusitania. The use of triazole antifungal agents has been associated with C krusei and C glabrata pathogens. The use of intravenous devices has been associated with outbreaks of C parapsilosis, C krusei, and C tropicalis.

Several reports have noted that the use of azole antifungal agents has been associated with a marked shift in Candida species.[16-20] The emergence of non-albicans species poses a challenge to clinicians, because they may be less susceptible to antifungal agents.[14,21]

Moulds

The other major group of fungal pathogens is moulds. Aspergillus is the most common mould found in cancer patients, representing approximately two-thirds of mould pathogens and 15% to 25% of all invasive fungal infections. Aspergillus (and other moulds) are not ordinarily found in or on the human body. Rather, they are most commonly acquired by inhalation of airborne organisms. The most prevalent species are A fumigatus and A flavus, with A niger, A terreus, and others being much less common.

Other Fungal Pathogens

In the last 2 decades, other formerly rare fungal organisms have emerged as opportunistic pathogens. Among these are Fusarium spp, Acremonium spp, Penicillium spp, dematiaceous moulds such as Pseudalleschera boydii, Bipolaris spp, Alternaria spp, Paecilomyces spp, Trichosporon spp, members of the Mucorales order, Trichophyton spp, Microsporum, and others. Descriptions of these new human pathogens can be found in the journals Transplant Infectious Diseases and Clinical Infectious Diseases.[21,22]

It is important to note that except for the genus Candida, infection of humans by most fungal species occurs "accidentally," with the human host not representing a significant reservoir of organisms. This means that in most instances fungal infections are not contracted by person-to-person contact, but rather acquired through exposure to a source in nature or the hospital environment.

Recovery of such nonresident organisms in a patient suspected of having an infection should, therefore, be of much greater concern than recovery of organisms that are ordinarily part of the body flora (eg, Candida) and have no clinical meaning per se. The clinical significance of fungal isolates from mucosal or external body surfaces as potential causes of invasive infection must be considered in the context of other factors, such as the degree of host compromise, the integrity of host barriers, and so forth, in order to interpret whether their presence supports a diagnosis of invasive fungal infection.