Dr. J. Souglakos was the recipient of a CABR clinical fellowship.
Nonneutropenic febrile episodes associated with docetaxel-based chemotherapy in patients with solid tumors
Article first published online: 5 SEP 2002
Copyright © 2002 American Cancer Society
Volume 95, Issue 6, pages 1326–1333, 15 September 2002
How to Cite
Souglakos, J., Kotsakis, A., Kouroussis, C., Kakolyris, S., Mavroudis, D., Kalbakis, K., Agelaki, S., Vlachonikolis, J., Georgoulias, V. and Samonis, G. (2002), Nonneutropenic febrile episodes associated with docetaxel-based chemotherapy in patients with solid tumors. Cancer, 95: 1326–1333. doi: 10.1002/cncr.10802
- Issue published online: 5 SEP 2002
- Article first published online: 5 SEP 2002
- Manuscript Accepted: 16 APR 2002
- Manuscript Revised: 22 JAN 2002
- Manuscript Received: 29 JAN 2001
- Cretan Association for Biomedical Research (CABR)
- solid tumors;
- nonneutropenic infections
Docetaxel is associated with severe lymphopenia and increased incidence of nonneutropenic infection. This study investigated the incidence of nonneutropenic infections and/or febrile episodes in patients with solid tumors receiving frontline docetaxel-based chemotherapy.
Chemotherapy-naive patients with solid tumors treated with docetaxel-based chemotherapy were studied prospectively for the development of nonneutropenic infections.
During a 2-year period, 680 cancer patients enrolled in 24 protocols received 2867 cycles of docetaxel-containing chemotherapy. Fifty-three patients (7.8%) developed nonneutropenic infections and/or febrile episodes. The most common of these were pulmonary infections (n = 25), Pneumocystis carinii interstitial pneumonias (n = 5), and candidal (n =11), herpetic (n =4), and cytomegaloviral (n =3) infections. Thirty-six patients (68%) had severe lymphopenia (< 900 cells per deciliter) and 49 (92%) had less than 400 CD4+ cells per deciliter. Patients with a low CD4+ cell count (≤ 200 cells per deciliter) had a significantly higher probability to develop opportunistic than common infections (P = 0.002). The incidence of nonneutropenic infections and/or febrile episodes was significantly higher in patients treated with docetaxel/gemcitabine (18.3%; P = 0.0001) and docetaxel/CDDP (11.7%; P = 0.012) than in those treated with docetaxel alone (3.6%). Conversely, 175 patients who received 752 cycles of chemotherapy with paclitaxel-containing regimens and 410 patients who received 2174 cycles with nontaxane-based regimens developed 6 (3.4%; p=0.042) and 12 (3%; p=0.001) nonneutropenic infections, respectively. Less than 10% of the patients of the two latter groups were lymphopenic. The risk of nonneutropenic infection in patients receiving docetaxel-based chemotherapy was 2.38 and 2.8 times higher than in patients receiving paclitaxel and nontaxane-based chemotherapy, respectively.
Patients treated with docetaxel-based chemotherapy are at increased risk of developing nonneutropenic infections. This may be related, at least partly, to severe postchemotherapy CD4+ lymphopenia. Cancer 2002;95:1326–33. © 2002 American Cancer Society.
Docetaxel is a semisynthetic taxoid. Its cytotoxic effect is based on its ability to promote microtubule assembly and to inhibit the depolymerization of tubulin, which subsequently leads to microtubule stabilization.1, 2 Docetaxel has a broad spectrum of antitumor activity as frontline chemotherapy against breast, ovarian, and nonsmall cell lung carcinomas (NSCLC).3–5 In addition, the drug has been used to treat patients with esophageal, gastric, pancreatic, head and neck carcinomas, and soft tissue sarcomas with less impressive but, still, interesting results.6–9
Docetaxel is associated with severe adverse events including alopecia, diarrhea, asthenia, hypersensitivity and cutaneous reactions, onycholysis, neurotoxicity, weakness, and weight gain due to fluid retention syndrome. The drug also causes severe Grade 3 and 4 leukopenia and neutropenia.3–9
Toxicity of the lymphohematopoietic system was observed in experimental animals and included lymphopenia, lymphoid depletion, involution of lymphoid organs, and thymic atrophy.10, 11 Effects on human lymphocytes have not been described in clinical trials. Our group reported a Phase I trial of docetaxel and mitoxantrone combination in chemotherapy-naive patients with advanced breast carcinoma. One of the main toxicities of this regimen was lymphopenia, which affected all lymphocyte subsets.12 In addition, in a prospective phenotypic study of 30 patients with advanced solid tumors who received single-agent docetaxel every 3 weeks as frontline chemotherapy, a statistically significant decrease of the absolute number of peripheral blood lymphocytes, i.e., CD3+, CD4+, CD56+ but not CD20+ cells, was observed after the administration of the first chemotherapy cycle. A second chemotherapy course resulted in a further decrease of the absolute number of all lymphocyte subsets including the CD20+ cells.13 This lymphopenic effect was not due to the corticosteroids that were used as premedication for docetaxel. This conclusion was reached because a group of cancer patients with metastases in the central nervous system who received only high-dose corticosteroids and radiotherapy as symptomatic therapy did not present similar quantitative abnormalities of the lymphocyte subsets.13 In addition, our group observed nonneutropenic infections in patients enrolled in Phase II trials evaluating different docetaxel-based chemotherapy regimens, many of whompresented with docetaxel-induced lymphopenia.14–20 Conversely, Reckzeh et al.21 reported severe lymphopenia and interstitial pneumonia in patients with NSCLCwho were treated with a combination of paclitaxel and radiotherapy.
Based on the above observations, we evaluated prospectively all patients receiving docetaxel-containing chemotherapy for the development of nonneutropenic febrile episodes. We also compared the incidence of nonneutropenic febrile episodes during docetaxel-based chemotherapy with the incidence of episodes in patients receiving paclitaxel and/or nontaxane-containing chemotherapy.
MATERIALS AND METHODS
All patients participating in the protocols performed at the Department of Medical Oncology, University Hospital of Heraklion (Crete, Greece) were studied prospectively for the development of infections during the period May 1996–May 1998. All infections were recorded and investigated to identify the nonneutropenic cases. After identification, the nonneutropenic febrile episodes were classified into three categories: febrile episodes occurring in patients receiving docetaxel-containing regimens; febrile episodes occurring in patients receiving paclitaxel-containing regimens; and febrile episodes occurring in patients receiving nontaxane-containing regimens. The two latter categories were used as controls, so that the incidence of the nonneutropenic febrile episodes associated with docetaxel-containing chemotherapy could be compared with that of the paclitaxel or nontaxane-containing chemotherapeutic combinations.
During the period under investigation, 680 patients received docetaxel, 175 received paclitaxel, and 410 received nontaxane-based regimens. No patient received prophylactic antibiotics. Treatment characteristics of the patients treated with docetaxel, paclitaxel, and nontaxane-based chemotherapy are shown in Tables 1, 2, and 3, respectively.
|Primary tumor||No. of febrile episodes|
|NSCLC (n = 345)|
|Docetaxel + RTX (35)||2|
|Breast carcinoma (n = 190)|
|Gastric carcinoma (n = 31)|
|Pancreatic carcinoma (n = 39)|
|Bladder carcinoma (n = 26)|
|Ovarian carcinoma (n = 4)|
|Cervical carcinoma (n = 3)|
|Soft tissue sarcomas-(n = 3)|
|Cholangiocarcinoma (n = 13)|
|Miscellaneous (n = 26)a|
|Total (n = 680)||53|
|Primary tumor||No. of febrile episodes|
|NSCLC (n = 29)|
|SCLC (n = 31)|
|Head and neck carcinoma (n = 13)|
|Cervical carcinoma (n = 16)|
|Bladder carcinoma (n = 5)|
|Ovarian carcinoma (n = 28)|
|Soft tissue sarcomas (n = 20)|
|Ca UP (n = 5)|
|Melanoma (n = 5)|
|Miscellaneous (n = 23)a|
|Total (n = 175)||6|
|Primary tumor||No. of febrile episodes|
|Colorectal carcinoma (n = 149)|
|Breast carcinoma (n = 122)|
|NSCLC (n = 28)|
|Hepatocellular carcinoma (n = 4)|
|Biliary tract carcinoma (n = 9)|
|Pancreatic carcinoma (n = 9)|
|Ovarian carcinoma (n = 2)|
|Bladder carcinoma (n = 6)|
|Soft tissue sarcomas (n = 38)|
|Renal carcinoma (n = 26)|
|Miscellaneous (n = 17)a|
|Total (n = 410)||12|
Each febrile episode was evaluated by medical history, physical examination, fever measurement, cultures of the infectious or suspicious material (e.g., pus, sputum, blood), chest X-rays, or computed tomographic scan and/or other imaging procedures if necessary. Bronchoscopy was performed in several cases of pulmonary infection upon appropriate clinical indications. Febrile episodes were characterized as clinically documented if the clinical picture and all other tests strongly suggested an infection but the causative organism was not isolated, and microbiologically documented if the causative organism was identified. All patients were evaluated clinically by two different investigators (J.S. and G.S.) from the medical staff. In case there was any disagreement on the clinical evidence of the infection, the event was not included in the study. A fever was documented as such if there was an increase in temperature above 38 °C, if it lasted more than 2 hours, and if it was unrelated to the administration of pyrogenic agents.
The rest of the workup consisted of complete hematologic and biochemical evaluation, including complete blood counts with differential and platelets every day during the febrile episode. CD4 and CD8 lymphocyte counts of peripheral blood were determined using appropriate monoclonal antibodies by indirect immunofluorescence and fluorescence-activated cell sorting analysis.22 Serum immunoglobulin (IgA, IgG, IgM) levels were also determined by nephelometry. Both lymphocyte subsets and serum immunoglobin levels were measured in all patients at the onset of the infectious episode.
Mild and moderate infections, such as those requiring only topical or oral therapy, if documented, were included in the study. Neutropenia was defined as a neutrophil count of 1000 or fewer cells per deciliter,lympocytopenia as a lymphocyte count of 900 cells or fewer per deciliter, CD4+ lymphopenia as a cell count of 400 or fewer cells per deciliter, and CD8 lymphopenia as a cell count of 300 cells or fewer per deciliter. Immunoglobulins were considered low if the level of IgA was 25 mg/dL or less, IgG was 700 mg/dL or less, and IgM was 40 mg/dL or less.
Statistical analysis was performed using the chi-square test for the analysis of the two-dimentional contingency table. Relative risks were approximated by odds ratios, calculated by standard methods.23
During the 2-year study period, 1265 patients received 5793 cycles of chemotherapy. Six hundred eighty patients were treated with 2867 cycles of docetaxel-containing regimens, whereas 175 patients were treated with 752 cycles of paclitaxel-containing regimens. Another 410 patients received 2926 cycles of nontaxane-containing chemotherapy treatment. All patients in the three study groups were studied and analyzed prospectively for the development of infections or febrile episodes.
Among the 680 and the 175 patients treated with docetaxel and paclitaxel-based chemotherapy, 53 (7.8%) and 6 (3.4%) developed nonneutropenic febrile episodes, respectively. The incidence of nonneutropenic febrile episodes observed in patients treated with docetaxel-based regimens was significantly higher than that observed in patients treated with paclitaxel-based regimens (P = 0.042). Similarly, among the 410 patients treated with nontaxane-containing regimens, 12 (3%) developed nonneutropenic febrile episodes. This incidence was also significantly lower than that observed in docetaxel-treated patients (P = 0.001).
The risk of developing nonneutropenic febrile episodes among patients treated with docetaxel-based regimens was 2.38 times higher than the risk of patients receiving paclitaxel-based regimens (95% confidence interval [CI: 1.5–3.2) and 2.8 times higher than the risk of patients receiving nontaxane-based chemotherapy (95% CI: 2.16–3.44). The same risk among patients treated with paclitaxel was 1.17 times higher than the risk among those treated with nontaxane-based chemotherapy (95% CI: 0.18–2.17).
Table 4 shows the types of infection observed in patients treated with docetaxel-based chemotherapy. Clinically documented infections of the lower respiratory tract, Pneumocystis carinii pneumonias (PCP), and candidal, cytomegaloviral(CMV), and herpetic infections were the most common infections seen in patients treated with docetaxel-containing regimens.
|Type of infection||No. of patients (%)|
|Pulmonary infection||25 (3.7)|
|Candidal infections||11 (1.5)|
|Pneumocystis carinii pneumonia||5 (0.7)|
|Herpes zoster||4 (0.6)|
|Cytomegalovirus esophagitis||3 (0.3)|
|Pulmonary abscess||1 (0.2)|
|Perianal abscess||1 (0.2)|
|Urinary tract infections||1 (0.2)|
Most febrile infections of the lower respiratory tract were diagnosed clinically and radiologically. In particular, 16 patients developed bronchopneumonias with radiologically detected, diffuse, patchy infiltrates and 9 patients had chest X-rays suggesting lobar consolidation. One patient had a chest X-ray that showed a cavity and an air-fluid level, suggesting a pulmonary abscess. Although these infections were documented clinically, definitive microbiologic identification of the causative organism was not achieved. All patients were treated with broad spectrum antibiotics that provided gram-positive, gram-negative, and anaerobic coverage. Five other patients with shortness of breath, hypoxia, reduced performance status (PS), and a particularly low CD4+ absolute lymphocyte number (< 200 cells per deciliter) developed a diffuse reticulonodular pattern on chest X-rays that was highly suggestive of PCP. Examination of the bronchoalveolar lavage after fiberoptic bronchoscopy confirmed the diagnosis and all patients were treated with high-dose trimethoprim/sulfamethoxazol and corticosteroids. One patient with lobar pneumonia, one with bronchopneumonia, and one with PCP died.
Two patients with fever and chills had blood stream infections. One of these patients developed Acinetobacter calcoaceticus, the other Staphylococcus aureus bacteremia. Both were treated appropriately with antimicrobial agents to which the causative organisms were sensitive. The patient suffering from A. calcoaceticus bacteremia finally went into shock and died.
One febrile patient with perianal pain developed a perirectal abscess due to a mixed gram-negative and anaerobic flora. The abscess was drained. The patient was treated successfully with antibiotics providing gram-negative and anaerobic protection.
Eleven patients developed microbiologically confirmed severe oropharyngeal and esophageal candidiasis. All were treated successfully with fluconazole. One patient with a high fever and chills developed multiple skin lesions, which were examined by skin biopsy and cultures, both of which revealed Candida albicans infection. The patient was treated successfully with high-dose fluconazole. One patient with severe dysphagia and odynophagia developed CMV esophagitis. This patient was diagnosed by esophagoscopy, mucosal biopsy, and histology and was treated successfully with ganciclovir. Another patient presented with a progressive increase in IgM anti-CMV antibody titers, acute hepatitis, and severe liver dysfunction. Polymerase chain reaction (PCR) detected CMV infection in her blood. She was also treated successfully with ganciclovir. A third patient developed symptoms and signs of generalized CMV disease (pneumonitis, hepatitis, and typical retinitis). She also had a progressive increase in IgM anti-CMV antibody titers. The viral infection was also detected by PCR in her blood. Despite appropriate treatment, the patient died. Four patients suffered from a painful herpes zoster infection, two of whom developed disseminated skin disease. All were treated with early initiation of acyclovir and had an uneventful course. Other nonneutropenic infections were rare (Table 4) and were treated successfully with appropriate antibiotics.
Table 5 demonstrates that 2.8% (P = 0.008), 3.8% (P = 0.081), and 1.7% (P = 0.06) of the docetaxel/gemcitabine, docetaxel/radiotherapy, and docetaxel/CDDP cycles, respectively, were complicated with nonneutropenic infections.
|Docetaxel-based regimen||Treated patients/nonneutropenic infections (%)||P valuea||Administered cycles/nonneutropenic infections (%)||P value|
|Docetaxel||112/4 (3.6)||334/4 (1.2)|
|Docetaxel/CDDP||123/9 (7.3)||0.012||527/9 (1.7)||0.060|
|Docetaxel/gemcitabine||185/18 (9.7)||0.0001||640/18 (2.8)||0.008|
|Docetaxel/anthracycline||115/11 (9.6)||0.069||629/11 (4.2)||0.628|
|Docetaxel/navelbine ± CDDP||110/7 (6.4)||0.338||632/7 (1.1)||0.900|
|Docetaxel/radiotherapy||35/4 (11.4)||0.074||105/4 (3.8)||0.081|
All 53 patients who developed docetaxel-associated infections had normal neutrophil and lymphocyte counts when they were enrolled in the corresponding chemotherapy protocols. None of the patients was neutropenic at the time of the infection. However, five patients had presented with Grade 3 or 4 neutropenia, without fever, during the 20 days preceding the nonneutropenic infections. Four of these patients developed infections of the lower respiratory tract and one developed generalized candidiasis during this study.
Of 20 patients with opportunistic infections, 13 (65%) presented with a low absolute lymphocyte number (< 900 cells per deciliter) at the time of infection, as did 17 (52%) of 33 patients with nonopportunistic infections. Moreover, 90% and 94% of the patients who developed opportunistic and nonopportunistic infections, respectively, displayed less than 400 CD4+ lymphocytes per deciliter (Table 6). Four patients had low serum levels of both IgG and IgA and all developed lower respiratory tract infections. Eight other patients had low IgG serum levels, three of whom developed pulmonary infections, one PCP, one CMV eosophagitis, one oral candidiasis, one disseminated candidiasis, and one herpes zoster infection.
|Type of infection||No. of patients||No. of patients (%)||Mean ± SD cells/dL|
|ALN < 900/dL (%)||CD4+ < 400/dL (%)||ALN||CD4+|
|Opportunistica||20||13 (65)||18 (90)||733 ± 367||193 ± 121|
|Nonopportunistic||33||17 (52)||31 (94)||886 ± 477||193 ± 151|
Among the 175 patients who received paclitaxel-containing regimens, only 3presented with severe (< 900 cells per deciliter) lymphopenia associated with a low absolute CD4 cell count (< 400 cells per deciliter). One patient developed a herpes zoster infection, one a generalized herpes infection, and the otheresophageal candidiasis. Similarly, among the 410 patients who received nontaxane-containing combinations, 3 developed severe lymphopenia (< 900 cells per deciliter) with a low absolute CD4 count (< 400 cells per deciliter). One developed a severe urinary tract infection, the second PCP, and the third herpes zoster infection. Patients treated with docetaxel-based chemotherapy regimens had significantly lower CD4 and CD8 counts than patients treated with either paclitaxel (P = 0.01 and P = 0.03, respectively) or nontaxane-containing regimens (P = 0.06 and P = 0.03, respectively).
Among the 680 patients who were treated with docetaxel-containing regimens, 45 (13%) had received previous radiotherapy, whereas 20 patients (11.4%) who received paclitaxel-containing and 30 patients (7.3%) who received nontaxane-based chemotherapy had received previous radiotherapy, respectively. Two patients who had been treated with docetaxel-containing regimens developed nonneutropenic infection, whereas none of the patients who had received paclitaxel-containing or nontaxane-based chemotherapy developed nonneutropenic infection.
In the stepwise multivariate logistic regression analysis, the type of treatment was predictive of the development of nonneutropenic infection. The effect of treatment was significant (P < 0.01) and the hazard ratio for infection was 2.4 times higher in the docetaxel group than in the nondocetaxel group, given the same type of neoplasia, stage, age, and PS. Patients presenting with an absolute number of less than 200 CD4+ cells per deciliter after receiving docetaxel-based chemotherapy had a significantly higher probability of developing an opportunistic than a common infection (P = 0.002).
The results of the current study demonstrate, for the first time, that treatment of patients with solid tumors with docetaxel-containing chemotherapeutic regimens was associated with a significantly higher incidence of nonneutropenic febrile episodes, compared with the incidence of nonneutropenic febrile episodes associated with paclitaxel or nontaxane-containing regimens. Indeed, 7.8% of patients treated with docetaxel-based chemotherapy developed nonneutropenic febrile episodes and 37.7% of them developed opportunistic infections. The risk of developing a nonneutropenic febrile episode in patients treated with docetaxel-based chemotherapy was 2.38 times higher than the risk of patients receiving paclitaxel-based regimens and two to eight times higher than the risk of patients treated with nontaxane-based chemotherapy. In addition, our data indicate that the development of nonneutropenic febrile episodes in this patient population was associated with a low absolute (< 900 cells per deciliter) and CD4+ (< 400 cells per deciliter) lymphocyte count.
Infections have been reported frequently in patients treated with docetaxel-based chemotherapy. In most cases, these infections complicate docetaxel-induced neutropenia.1–7, 14–19 The high incidence of infections observed in the current study cannot be attributed to chemotherapy-induced neutropenia24 because all patients had neutrophil counts within normal limits at the time of infection. It is possible that patients were infected at an earlier time, when they were neutropenic due to docetaxel-containing chemotherapy, but did not develop the symptoms of the infection until later, i.e., after they had recovered from the neutropenia. However, only 5 of 53 patients presented with a Grade 3 or 4 neutropenic episode during the 20 days preceding the development of the nonneutropenic infection. This observation strongly suggests that the increased risk for the development of nonopportunistic febrile episodes is not related directly to a previous Grade 3 or 4 neutropenic episode. Conversely, 57% of the patients had less than 900 lymphocytes per deciliter, whereas 92% of them had less than 400 CD4+ cells per deciliter in their peripheral blood during the nonneutropenic infection. Furthermore, in the stepwise multivariate logistic regression analysis that was performed, the type of treatment was the best predictive factor for the development of nonneutropenic febrile episodes.
To the best of our knowledge, no study has evaluated the risk of nonneutropenic febrile episodes in patients treated with taxane-based regimens. Reckzeh et al.21 described severe lymphopenia and interstitial pneumonia in 14 patients treated with paclitaxel and simultaneous radiotherapy for NSCLC. In our study, only two patients who had received previous radiotherapy and docetaxel-based chemotherapy had developed a nonneutropenic infection. One of these patients had a microbiologically proven PCP infection. The authors have shown that treatment with paclitaxel was associated with a low absolute number of lymphocytes and a low CD4+ cell count.21 In our study, a number of patients in the docetaxel, paclitaxel, and nontaxane regimens had received previous radiotherapy. Our group reported nonneutropenic febrile episodes in patients with various solid tumors who were treated with docetaxel-based chemotherapy in Phase II trials.14, 16–18 We also reported severe lymphopenia with decreased CD4+ and CD8+ cell counts associated with two cases of interstitial pneumonia among patients with NSCLC treated with docetaxel and concomitant radiotherapy.20 In addition, we reported severe lymphopenia in patients with metastatic breast carcinoma who were treated with frontline docetaxel in combination with mitoxantrone. In that study, the lymphocyte depletion was not restricted to any specific cell subset.12 In addition, we reported that even one cycle of docetaxel monotherapy, administered as frontline treatment with solid tumor, included a significant decrease in the absolute number of peripheral blood CD3+, CD4+, CD8+, and CD56+ but not CD20+ lymphocytes.13 Our findings indicate that docetaxel-containing regimens may induce severe lymphocytotoxicity in humans, which may be complicated by a high incidence of nonneutropenic infections and/or febrile episodes.
It is noteworthy that the incidence of nonneutropenic febrile episodes was higher in patients treated with docetaxel-based combinations than in patients who received single-agent docetaxel (Table 3). The higher incidence of nonneutropenic febrile episodes was observed in patients treated with docetaxel/gemcitabine and docetaxel with concomitant radiotherapy.24 The reasons of this effect are still unclear. However, nonneutropenic febrile episodes were also observed with different docetaxel-based combinations, but the small number of observed cases precludes a valuable statistical analysis. Additional prospective studies are needed to define whether the combination of docetaxel with certain anticancer agents is more immunosuppressive and can cause a higher incidence of nonneutropenic febrile episodes.
The purine analogs fludarabine and 2-chlorodeoxyadenosine are anticancer drugs with a well known toxicity profile that includes severe immunodeficiency. This immunodeficiency is due to a protracted lymphopenia with a low absolute number of T cells, mostly, but not exclusively, CD4+ cells. In all reported cases, the time for complete recovery of the lymphatic system is longer than 3 months.25, 26 One study analyzed infections in patients with chronic lymphocytic leukemia treated with fludarabine.27 Anaissie et al. showed that this agent, when used in previously treated patients, may be associated with infections due to T-cell dysfunction, such as listeriosis, pneumonocystosis, mycobacterial infections, as well as opportunistic fungal and viral infections. In the current study, five patients had documented PCP infection. Eleven patients had candidiasis (one of them had a generalized infection), four had herpes zoster infection (two of them generalized), and three patients had CMV infections (two of them generalized). All these patients had a very low absolute number of CD4+ cells. It is well known that T-lymphocyte dysfunction is the main predisposing factor for the development of fungal, viral, and parasitic infections.26, 27 Our finding that the risk of developing opportunistic infections was significantly higher in patients with less than 200 CD4+ cells per deciliter supports this hypothesis. Therefore, it is reasonable to hypothesize that severe docetaxel-induced CD4+ lymphopenia is the main predisposing factor for the development of nonneutropenic opportunistic infections observed in these patients.
An association between CD4+ lymphopenia and febrile nonneutropenic episodes has been demonstrated. However, additional factors (e.g., type and location of the tumor, chemotherapy-induced tissue and membrane damage, alterations of the microbial flora, or the intensity of specific regimens) may have contributed to the development of nonneutropenic fever and infection. Given the results of the current study, additional investigations are required to evaluate the relative contribution of these factors to the development of febrile episodes. Conversely, only a few patients had low IgG and IgM levels. Therefore, an association between immunoglobulin levels and the described infectious episodes could not be shown.
The current data demonstrate a higher incidence of fever with the docetaxel-based combinations. This is not necessarily due to the docetaxel itself, but to the combination of docetaxel with other drugs. Although the cause and effect relationship has not been established definitiveely, the CD4+ data strongly suggest that the immunosuppression associated with the docetaxel-based combinations may be implicated.13 However, other coexisting factors predisposing to febrile episodes such as the additive toxic effects by other drugs to the mucous membranes or other tissues, e.g., the lungs28 or the intestine,29 cannot be excluded. Because growth factors were not used in most of these nonneutropenic episodes, a paradoxical effect of fever due to growth factor usage cannot be implicated.
In conclusion, docetaxel-containing treatments may be complicated by severe CD4+ lymphopenia. This may result in immunodeficiency, predisposing patients to the development of both opportunistic and nonopportunistic infections and/or febrile episodes. Therefore, patients treated with docetaxel-based chemotherapy and clinically apparent infection should be monitored carefully and appropriate treatment should be administered as early as possible.
- 4Summary of phase II data of docetaxel (Taxotere), an active agent in the first-and second-line treatment of advanced non-small cell lung cancer. Semin Oncol. 1995; 22 (Suppl 4): 22–29., , , et al.
- 9Phase II studies of docetaxel in the treatment of various solid tumors. Eur J Cancer. 1993; 31A (Suppl 4): S2–4., , , et al.
- 17Docetaxel (Taxotere) and vinorelbine (Navelbine) in the treatment of non-small cell lung cancer. Semin Oncol. 1997; 24 (suppl 14): 9–14., , , et al.
- 18Docetaxel (Taxotere) and gemcitabine (Gemsar) in the treatment of non-small cell lung cancer: preliminary results. Semin Oncol. 1997; 24 (Suppl 14): 22–25., , , et al.
- 23Statistical methods in clinical research, 2nd ed. Oxford: Blackwells, 1994, .
- 28A phase I study of weekly docetaxel(D) and gemcitabine (G) in advanced non-small cell lung cancer [abstract]. International Congress on Lung Cancer, Rhodes, Greece, 2001:152., , , et al.