Weekly docetaxel is safe and effective in the treatment of advanced-stage acquired immunodeficiency syndrome-related Kaposi sarcoma


  • Soon Thye Lim M.D.,

    1. Department of Hematology, University of Southern California Keck School of Medicine/Norris Comprehensive Cancer Center, Los Angeles, California
    Current affiliation:
    1. Department of Medical Oncology, National Cancer Center of Singapore, 11 Hospital Drive, Singapore 169619, Singapore
    Search for more papers by this author
  • Anil Tupule M.D.,

    1. Department of Hematology, University of Southern California Keck School of Medicine/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Byron M. Espina B.Sc.,

    1. Department of Hematology, University of Southern California Keck School of Medicine/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Alexandra M. Levine M.D.

    Corresponding author
    1. Department of Hematology, University of Southern California Keck School of Medicine/Norris Comprehensive Cancer Center, Los Angeles, California
    • University of Southern California Keck School of Medicine/Norris Comprehensive Cancer Center, 1441 Eastlake Avenue, Los Angeles, CA 90033
    Search for more papers by this author
    • Fax: (323) 865 0060



Intravenous paclitaxel, 100 mg/m2, given over 3 hours every 2 weeks is associated with a response rate of 59% in patients with recurrent or refractory acquired immunodeficiency syndrome (AIDS)-related Kaposi sarcoma (KS). However, this regimen is associated with significant myelosuppression, and the inconvenience of a 3-hour infusion. Moreover, no effective therapies have been defined for use after treatment failure with this agent. A Phase II trial was conducted with weekly docetaxel in patients with advanced-stage KS to assess safety and antitumor activity.


Docetaxel was administered at a dose of 25 mg/m2 intravenously over 15–30 minutes weekly for 8 weeks. Thereafter, if the patient experienced stable disease or better response, treatment doses were given every other week until complete disease remission, disease progression, or unacceptable toxicity occurred.


Twelve patients were accrued—9 had > 25 mucocutaneous lesions, 1 had lymphedema, and 2 had visceral involvement. Ten patients (83%) had previous systemic chemotherapy, including 4 who received previous paclitaxel. Treatment was well tolerated, with no Grade 4 toxicity of any type. Grade 3 neutropenia occurred in 33% of patients but no patient had neutropenic fever. Five patients (42%) achieved a partial response, including 1 who had previously failed to respond to paclitaxel. The median time to disease progression was 26 months (range, 5–53 months). With a median follow-up period of 45 months, the median survival point had not been reached.


Weekly docetaxel is safe, with reasonable antitumor activity in patients with advanced-stage, recurrent, or refractory AIDS-related KS. Cancer 2005. © 2004 American Cancer Society.

Kaposi sarcoma (KS) is the most common neoplasm associated with human immunodeficiency virus (HIV) infection. In the past decade, a number of treatment modalities have been explored for acquired immunodeficiency syndrome (AIDS)-related KS. Cytotoxic chemotherapy remains the most successful modality in the treatment of advanced-stage KS. Currently, liposomal doxorubicin and daunorubicin are both licensed for use in patients with advanced-stage KS, with response rates ranging from 30% to 70%.1–3 Despite this high response rate, cessation of chemotherapy is nearly always associated with a rather rapid recurrence of disease.

Paclitaxel has shown significant efficacy as a second line treatment for AIDS-related KS, and is approved for this use in the United States. We have reported previously our experience with biweekly paclitaxel administered at a dose of 100 mg/m2.4 The regimen was associated with a response rate of 59% and a median survival period of 15.4 months. However, myelosuppression was significant and there was a relatively high prevalence of alopecia, myalgia, and fatigue. These toxicities, coupled with the need for a 3-hour infusion, makes paclitaxel less than optimal for patients with AIDS-related KS.

Docetaxel, a member of the taxane family, previously has been studied in patients with AIDS-related KS.5 Forty-one patients with advanced-stage KS were treated with docetaxel at a dose of 60 mg/m2 intravenously (i.v.) every 3 weeks. All patients had received at least one previous regimen of chemotherapy and 21 (52%) had previous exposure to an anthracycline. Partial responses (PR) were reported in 18 of the 27 evaluable patients (67%), which occurred after a median of 3 cycles. However, despite the reduced dose of docetaxel, and the finding that all patients received granulocyte—colony-stimulating factor (G-CSF) support, 27% of patients still experienced Grade 4 neutropenia.

Subsequent experience with docetaxel in solid tumors indicated that weekly use of the drug at lower doses might maintain efficacy while reducing toxicity. In a Phase II study involving patients with advanced-stage breast carcinoma, weekly docetaxel at 36 mg/m2 for 6 consecutive weeks followed by 2 weeks without treatment was both efficacious and well tolerated. Severe neutropenia occurred in only 0.4% of courses and no other hematologic toxicity was observed.6

We hypothesized that weekly docetaxel would be effective, with less toxicity in the treatment of patients with recurrent or refractory AIDS-related KS. To test this hypothesis, we conducted a Phase II trial of docetaxel at a dose of 25mg/m2 given weekly for 8 weeks, followed by every other week dosing thereafter.


Study Criteria

Twelve patients with AIDS-related KS were accrued from March 2000 to November 2001. Written informed consent was obtained from all patients, after approval of the study by our institutional review board (IRB). Patients were required to have serologic evidence of HIV-1 infection, with histologically confirmed KS. Advanced-stage KS was required for study entry, including either ≥ 25 mucocuatneous lesions, or symptomatic visceral involvement, and/or KS-related lymphedema. Previous therapy for KS with local or systemic agents was allowed. Other eligibility criteria included the following: age ≥ 18 years, Karnofsky performance status ≥ 50%, adequate bone marrow function (i.e., absolute granulocyte count >1000/mm3 and platelet count > 75,000/mm3), creatinine level < 2.1 mg/dL, and adequate hepatic function (i.e., aspartate aminotransferase [AST] and alanine aminotransferase level < 3 times normal and bilirubin level < 2.0 mg/dL). Use of G-CSF was allowed.

Treatment Plan

Docetaxel was administered at a dose of 25 mg/m2 i.v. over 15–30 minutes weekly for 8 weeks. Thereafter, if the patient experienced stable disease (SD) or better response, additional treatment was administered at the same dose, but every other week. Doses were given provided hematologic function was adequate (absolute neutrophil count > 1000/mm3, platelet count > 50,000/mm3). Growth factor support was given as needed. Patients were premedicated with 8 mg dexamethasone orally twice daily for 3 days, starting 24 hours before each docetaxel infusion.

During protocol therapy, dose adjustments were made. The dose of docetaxel was reduced by 20% when the alkaline phosphatase (ALP) level was elevated but was < 5 times the upper limit of normal (ULN). It was also reduced by 20% when the AST level is 1.6–5 times the ULN. However, if the ALP or AST level was ≥ 5 times the ULN, or if the bilirubin level was greater than the ULN, docetaxel administration was delayed for ≤ 3 weeks. After normalization of these values, docetaxel was administered at a 20% dose reduction. For the first episode of Grade 2 peripheral neuropathy, docetaxel was withheld until the toxicity decreased to ≤ Grade 1. Docetaxel was then resumed at 100% of the dose. If a second episode of Grade 2 neuropathy occurred, or if the neuropathy was persistently ≥ Grade 2, the patient was removed from the study.

Patients continued to receive treatment until they developed complete disease remission, disease progression, recurrent Grade 3/4 toxicity (except hematologic toxicity), or an intercurrent opportunistic infection.


The baseline assessment included a history and physical examination, a complete blood count, comprehensive blood chemistry evaluations, a CD4 count, and chest X-ray (CXR) and other evaluations, as clinically indicated. Appropriate endoscopic procedures were performed in patients with symptoms suggestive of visceral gastrointestinal involvement. Up to 25 mucocutaneous lesions were counted and characterized as flat or raised. In the presence of > 25 lesions, 1 anatomic site was selected and followed for disease parameters. In addition, up to five indicator lesions were measured in two dimensions and the product of the two diameters recorded. Tumor measurements were obtained every 4 weeks for the first 12 weeks, and then every 8 weeks thereafter. Diagnostic endoscopic procedures were also repeated to document response if initially positive.

The AIDS Clinical Trials Group (ACTG) response criteria were used for assessment of tumor response.7 A complete response (CR) was defined as resolution of all disease for ≥ 4 weeks without evidence of new disease. In patients with residual pigmentation, all lesions must have been completely flat. In addition, a repeat biopsy and pathologic evaluation must have demonstrated no evidence of disease in a representative lesion. Patients with visceral involvement must have received a repeat endoscopic or other specific procedure to document the absence of visceral disease. A PR was defined as a ≥ 50% decrease in the number, bidimensional measurement, and/or nodularity of previously existing lesions (e.g., skin, oral, measurable disease) lasting for ≥ 4 weeks, without new skin or oral lesions or appearance of tumor-associated edema or visceral involvement. Patients with visceral involvement were required to demonstrate a ≥ 50% reduction of tumor parameters compared with baseline. Progressive disease (PD) was defined as the development of new lesions, an increase in the bidimensional measurements by ≥ 25%, a change in 25% of previously flat lesions to raised lesions, development of new or worsening of existing tumor-associated edema, or development of new or progressive visceral disease. SD was defined as not fulfilling criteria for a CR, a PR, or PD.

The onset of response was measured from the date of first therapy to the date of first recorded documentation of response. Time to disease progression was measured from the date of initial response to the date of disease progression. All patients meeting the eligibility criteria and receiving any therapy were evaluated for toxicity according to the National Cancer Institute Common Toxicity Criteria (Version 2.0).

Statistical Analysis

The current study was designed as a nonrandomized Phase II trial. All patients were considered in the analysis of response, on an intention-to-treat basis.


Clinical Characteristics

Twelve patients with AIDS-related KS were enrolled. The clinical characteristics are listed in Table 1. All patients were men, with a median age of 38 years (range, 27–49 years). All patients received highly active antiretroviral therapy (HAART). The median duration of previous protease inhibitor use was 12.4 months (range, 1.0–52.2 months). All patients had at least one of the poor risk factors for response as defined by ACTG criteria. Nine patients had > 25 mucocutaneous lesions, 1 patient had lymphedema, and 2 patients had visceral involvement that included the gastrointestinal tract.. In addition, three patients had a previous AIDS-defining opportunistic infection. All except 2 patients (83%) had received previous systemic chemotherapy for KS and 4 patients received at least two previous regimens of systemic chemotherapy.

Table 1. Demographics and Baseline Characteristics (n = 12)
CharacteristicsNo. of patients (%)
  1. AIDS: acquired immunodeficiency syndrome; HIV: human immuno deficiency virus; HAART: highly active antiretroviral therapy; KS: Kaposi sarcoma; GI: gastrointestinal.

Median age (range)38 (27–49 yrs)
Male gender12 (100)
Median Karnofsky performanance status 
 90%6 (50)
 80%4 (33)
 70%2 (17)
Previous AIDS-defining opportunistic  infection3 (25)
Median CD4+ count157 (1–402 mm3)
HIV RNA viral load 
 Not detectable2 (17)
 Median, range47,129 (623–225, 402 copies/mL)
Median previous protease inhibitor use12.4 (range, 1.0–52.2 mos)
Concurrent HAART12 (100)
Extent of KS 
 > 25 mucocutanoeus lesions9 (75)
 Lymphedema8 (67)
 Visceral involvement (GI)2 (17)
Previous systemic chemotherapy10 (83)
 Previous paclitaxel4
 Previous liposomal daunorubicin4
 Previous doxorubicin8


As shown in Table 2, treatment was well tolerated, with no Grade 4 toxicity of any type. Although Grade 3 neutropenia occurred in 33% of patients, no patient had neutropenic fever. Other hematologic toxicity included 1 patient (8%) with Grade 3 thrombocytopenia. Serious nonhematologic toxicity (Grade 3) was uncommon, occurring in only 1 patients(8%). The most common treatment-related Grade 1/2 nonhematologic toxicities were fatigue, myalgia, transaminitis, nail bed changes, insomnia, and neuropathy. Although 4 patients had elevated levels of liver enzymes, only 1 patient experienced Grade 3 elevation. Nail bed discoloration and pain occurred in 3 patients after a median of 17.3 (range, 10–27) weekly doses of docetaxel. This adverse event resulted in 2 patients who refused further chemotherapy—one after 10 doses and the other after 7 doses.

Table 2. Toxicities (n = 12)
ToxicitiesGrade 1/2Grade 3Grade 4Total (%)
 Neutropenia0404 (33)
 Thrombocytopenia0101 (8)
 Transaminases3104 (33)
 Fatigue4004 (33)
 Myalgia4004 (33)
 Nausea3003 (25)
 Nail bed changes3003 (25)
 Insomnia3003 (25)
 Neuropathy3003 (25)
 Emesis2002 (17)
 Mucositis2002 (17)
 Headache2002 (17)
 Alopecia2002 (17)
 Cellulitis1001 (8)


A median of 10 (range, 1–27) doses of docetaxel were given (Table 3). Two patients (17%) left the study after 1 dose of docetaxel and were not evaluable for response. Both patients had prolonged Grade 2 nausea and emesis, and declined further doses of docetaxel. In an intent-to-treat analysis, 5 patients (42%) had a PR, 4 patients (33%) had SD, and 1 patient (17%) had PD. Of the 5 patients who attained a PR, 3 had a history of previous protease inhibitor use for ≥ 18 months before study entry, whereas 2 received protease inhibitor therapy and docetaxel simultaneously. Of the four patients who had previously received paclitaxel, one patient attained a PR and two patients attained SD. The median time to disease progression was 26 months (range, 2–53 months). With a median follow-up period of 25 months, the median survival has not been reached.

Table 3. Previous Therapies and Response to Docetaxel
Patient no.No. of “docetaxel” “doses given”Duration of “previous” protease “inhibitor use”Previous “chemotherapy”Response to “docetaxel”
12715.3 mosPaclitaxelStable disease
21033.9 mosVinblastine, doxorubicinStable disease
   Daunorubicin, paclitaxel 
31010.4 mosDaunorubicin, doxorubicinStable disease
4134 daysNonePartial response
5247.8 mosDaunorubicinPartial response
61552.2 mosDoxorubicinPartial response
71010.9 mosNoneStable disease
893 wksPaclitaxel, doxorubicinPartial response
9912.4 mosPaclitaxel, doxorubicin, daunorubicinNot evaluable
1037.0 mosDoxorubicinProgressive disease
1116 mosDoxorubicinNot evaluable
12518 mosDoxorubicinPartial response


The current study demonstrates that low-dose, weekly docetaxel has clear activity in patients with advanced-stage AIDS-related KS, resulting in a 42% response rate, on this intent-to-treat analysis. This overall response rate is comparable to that reported with liposomal doxorubicin.1,2 Further, docetaxel also appeared to have activity in patients who had failed to respond to previous paclitaxel, resulting in a PR in one of four such patients, and SD in two. This finding is consistent with previous studies in patients with breast and ovarian carcinoma, in whom docetaxel was also effective in paclitaxel-resistant disease.8, 9

Use of HAART, with or without protease inhibitors, has been associated with objective tumor response in patients with AIDS-related KS, presumably by augmenting the immune status of the host, with subsequent immune control of the underlying human herpes virus 8 infection and KS.10, 11 In the current study, all patients received HAART therapy at the time that docetaxel was started, with a median duration of protease inhibitor therapy of 12.4 months. Of 5 patients with KS who attained a PR, 3 had received a stable regimen of protease inhibitor-containing HAART for ≥ 18 months before study onset, whereas 2 had started HAART therapy more recently. It is conceivable that the positive tumor response in these two patients was related to the relatively recent onset of HAART therapy. Nonetheless, prolonged protease inhibitor-containing HAART regimens did not control the KS in five additional patients, and three of five objective responders had received prolonged, stable regimens of HAART, with progressive KS at the time that docetaxel was initiated.

The toxicity profile of docetaxel compares favorably with our previous experience with paclitaxel, in which the incidence of Grade 3/4 neutropenia was 25% and 36%, respectively.4 In contrast, Grade 3 neutropenia was experienced by 33% of the patients in the current study and there was no patient with Grade 4 neutropenia. In addition, no patient experienced neutropenic fever. Nonhematologic toxicity was generally mild. However, nail bed changes and pain, which occurred in 3 patients (25%), proved to be a troublesome adverse event. Two of these patients declined further chemotherapy because of this adverse effect, one after 10 doses, and the other after 7 doses. Taxanes, in particular docetaxel, are well known to cause nail changes, especially with prolonged use on a weekly basis.12. Although not a life-threatening complication, this may affect the patient's quality of life.

The current study was initiated at a time when widespread use of HAART had already occurred in the United States, resulting in a remarkable decline in the incidence of KS, which has continued to the present time.13–15 Patient enrollment was restricted, and a decision was made to close the trial secondary to this poor accrual. Nonetheless, despite the few patients accrued, 12 patients completed the study, demonstrating clear evidence of the efficacy of docetaxol in this setting. Docetaxel had a better toxicity profile than we had previously observed with the use of paclitaxel in the same patient population.

In summary, we have shown that docetaxel has reasonable antitumor activity in patients with advanced-stage AIDS-related KS who failed previous regimens of chemotherapy, including paclitaxel. None of the patients who received docetaxel developed Grade 4 hematologic toxicity, and the drug was well tolerated when given weekly at doses of 25 mg/m2. Future studies employing even lower doses, or doses given less frequently, are warranted, in an attempt to avert the nail bed changes that may occur.