Successful carboplatin desensitization in patients with proven carboplatin allergy

Authors

  • Ronit Confino-Cohen M.D.,

    Corresponding author
    1. The Allergy and Clinical Immunology Unit, Meir General Hospital, Kfar Saba, affiliated with Sackler School of Medicine, Tel-Aviv University Tel-Aviv, Tel-Aviv, Israel
    • Allergy and Clinical Immunology Unit, Meir General Hospital, 44281 Kfar Saba, Israel
    Search for more papers by this author
    • Fax: (011) 972-9-7472362

  • Amiram Fishman M.D.,

    1. Gynecologic Oncology Unit, Meir General Hospital, Kfar Saba, affiliated with Sackler School of Medicine, Tel-Aviv University Tel-Aviv, Tel-Aviv, Israel
    Search for more papers by this author
  • Marco Altaras M.D.,

    1. The Allergy and Clinical Immunology Unit, Meir General Hospital, Kfar Saba, affiliated with Sackler School of Medicine, Tel-Aviv University Tel-Aviv, Tel-Aviv, Israel
    Search for more papers by this author
  • Arnon Goldberg M.D.

    1. The Allergy and Clinical Immunology Unit, Meir General Hospital, Kfar Saba, affiliated with Sackler School of Medicine, Tel-Aviv University Tel-Aviv, Tel-Aviv, Israel
    Search for more papers by this author

Abstract

BACKGROUND

Carboplatin is one of the most useful and well tolerated cytotoxic drugs for gynecologic malignancies. Hypersensitivity to carboplatin is not rare among patients receiving multiple recurrent treatments with this drug. The aim of the current study was to offer a safe and convenient carboplatin desensitization strategy to patients with a proven allergic reaction to this drug.

METHODS

Patients with an immediate objective allergic reaction to carboplatin were skin tested with the drug. A 6-hour carboplatin desensitization protocol was administered to the patients with a carboplatin-positive skin test on each of the following treatment courses.

RESULTS

Twenty-three patients with an allergic reaction to carboplatin and a positive skin test were included in the current study. Twenty patients (86.9%) were desensitized. One patient developed a mild urticarial rash. Nineteen patients tolerated 80 desensitization courses uneventfully.

CONCLUSIONS

The data presented a successful desensitization protocol for individuals with a proven allergic reaction to carboplatin. The protocol was safe and convenient and offered an effective therapeutic strategy to patients who required this drug. Cancer 2005. © 2005 American Cancer Society.

Carboplatin has been the mainstay chemotherapy of gynecologic malignancies over the past decade. It is usually well tolerated, easily administered, and associated with mild side effects.1 Nevertheless, with the extensive use of carboplatin, it has become evident that recurrent courses of the drug expose patients to the risk of developing hypersensitivity reactions.2–6 These reactions can be either mild cutaneus eruptions or more serious anaphylactic symptoms such as hypotension, bronchospasm, and cardiovascular collapse.7 The current data suggest that most of the reactions are immunologically mediated. The symptoms tend to appear after four to six treatment courses and persist in further treatments.7, 8 Most of them occur during or immediately after infusion of the drug. The symptoms, which are anaphylactic in nature and the skin tests with carboplatin that are usually positive, suggest an immediate immunoglobulin E (IgE)mediated reaction. Different management approaches have been suggested to solve this adverse reaction of carboplatin treatment. They include avoiding use of the drug in future chemotherapy, reintroducing carboplatin with premedication,8 and using a desensitization protocol.9–12 We present a large group of patients with gynecologic malignancies who have developed allergic reactions to carboplatin as manifested by the clinical symptoms and proved by a positive skin test.As carboplatin was essential for these patients, they underwent a desensitization protocol with carboplatin. Most patients tolerated it uneventfully and the drug could be administered thereafter, in the following treatment courses, with the same protocol without any symptom of hypersensitivity.

MATERIALS AND METHODS

Patients

During the years 1996–2004, 228 patients with ovarian carcinoma or primary peritoneal carcinoma and 26 patients with serous papillary carcinoma of the endometrium were treated in the Gynecologic-Oncology Unit, Meir General Hospital (Kfar-Saba, Israel) with either carboplatin as a single agent or a carboplatin-based combination chemotherapy regimen. Of the 228 patients with ovarian carcinoma, 112 had recurrence of the disease. Fifty-seven of them received carboplatin or carboplatin-based chemotherapy as second-line chemotherapy.

Those who developed objective symptoms of hypersensitivity reaction during or immediately after the infusion of carboplatin were further evaluated by skin tests. If skin tests were positive, a desensitization protocol was offered for the subsequent treatment courses.

Objective symptoms of hypersensitivity included diffuse erythema, urticaria, angioedema, hypotension (systolic pressure < 80 mm Hg), wheezing, and bronchospasm. Subjective symptoms of pruritus, shortness of breath, abdominal pain, and general discomfort were considered to be hypersensitivity symptoms only if they were combined with at least 1 objective symptom. Reactions were divided into two groups: mild, and moderate to severe, according to recently published grading.13 Mild reactions included all cutaneous reactions, not accompanied by symptoms affecting other organ systems. Moderate to severe reactions included reactions involving one or more body system, with or without skin involvement. Eleven patients developed delayed reactions (appearing > 3 hours after the infusion). The reactions included mild rashes, pruritus, abdominal pain, or discomfort. These patients were not included in the current study.

Skin Tests

All suspected hypersensitive patients were skin tested after informed consent was obtained. Intradermal skin tests were performed with carboplatin at concentrations of 0.1 and 1 mg/mL as previously described.9 Histamine and saline served as positive and negative controls, respectively. Skin tests were read 20 minutes after the intradermal injection. They were considered positive if the wheal diameter was > 5 mm compared with the negative control and had a surrounding flare.

Desensitization Protocol

A 6-hour desensitization protocol included 4 carboplatin solutions. The first 3 infusion bags contained 1/1000, 1/100, and 1/10 of the total dose diluted in 150 mL of D5W solution (Teva Medical, Ashdod, Israel). The last bag contained the remainder of the undiluted drug dose. Each dilution was delivered in drip infusion over 90 minutes, starting with the 1/1000 solution. The next higher concentration was delivered immediately after successful completion of the preceding infusion. Patients were premedicated with antiemetics (ondansetron hydrochloride) and dexamethasone as indicated for routine carboplatin chemotherapy. The same protocol was used in all future treatment courses with carboplatin.

RESULTS

Twenty-three patients treated with carboplatin for gynecologic malignancy were included in the current study. Their median age was 54.8 years (range, 37–76 years). All had objective symptoms of a hypersensitivity reaction and demonstrated a positive skin test to carboplatin. Eight patients (34.7%) had mild reactions and 15 patients (65.2%) had moderate to severe reactions involving ≥ 2 organ systems (Table 1).

Table 1. Clinical Characteristics of the Patients and the Adverse Reactions
CharacteristicsNo. of patients (n = 23) (%)
  • a

    Mild reactions included all cutaneous reactions not accompanied by symptoms of other body systems.

  • b

    Moderate to severe reactions included all reactions involving one or more organ systems, with or without skin involvement.

Mean age (yrs) (range)54.8 (37–76)
Diagnosis 
 Ovarian carcinoma15 (65.2)
 Primary peritoneal carcinoma6 (26.1)
 Endometrial carcinoma2 (8.7)
Hypersensitivity reaction 
 Milda8 (34.7)
 Moderate to severeb15 (65.2)

As demonstrated in Table 2, hypersensitivity reactions appeared relatively late along the carboplatin treatment (mean, 9 treatments; range, 2–17 treatments). Seventeen (74%) patients developed their hypersensitivity reaction during the second or third line of carboplatin treatments after 1 or 2 prolonged no-treatment intervals while disease remission had been achieved. There was no correlation between the severity of the initial hypersensitivity reaction and the number of the previous treatment courses or the number of prolonged no-treatment intervals (data not shown).

Table 2. Timing and Severity of Carboplatin Hypersensitivity Reaction
No. of carboplatin courses before the allergic reactionNo. of patients (%)
≤ 55 (21.7)
> 618 (78.3)
Severity of the allergic reactionMean no. of carboplatin courses before the allergic reaction
Mild9.3
Moderate to severe8.8

Skin tests to carboplatin were positive in all patients.

After the initial reaction and skin testing, 20 (86.9%) patients continued chemotherapy with carboplatin according to the desensitization protocol (Table 3). For two of three patients who were not desensitized, carboplatin was discontinued due to progression of their malignancy. One patient was lost to follow-up.

Table 3. Desensitization Protocol
  1. i.v.: intravenously; D5W: dextrose 5% in water.

Premedication:
 Dexamethasone: 8–12 mg (i.v.) single dose immediately before the infusion of carboplatin
 Ondansetron hydrochloride: (i.v.) single dose immediately before the infusion of carboplatin
Carboplatin—each solution diluted in 150 mL D5W
 1/1000 of the total dose—in 150 mL D5W over 1.5 hrs. If tolerated:
 1/100 of the total dose—in 150 mL D5W over 1.5 hrs. If tolerated:
 1/10 of the total dose in—150 mL D5W over 1.5 hrs. If tolerated:
The remainder of the dose in—150 mL D5W over 1.5 hrs.

There was no difference in the severity of the initial hypersensitivity reaction between the desensitized and the women who were not treated.

Of 81 desensitization treatments, 80 were tolerated uneventfully. In one patient, the first desensitization treatment was discontinued because of mild urticarial reaction and further carboplatin treatment was not attempted.

Nineteen of 20 women (95%) tolerated well 80 desensitization treatments (mean, 4.05 desensitization treatments per patient; range, 1–13 desensitization treatments per patient). No reductions in the chemotherapy dose were required. No additional premedication was administered.

DISCUSSION

Carboplatin therapy is the standard treatment for ovarian and other gynecologic malignancies. Although the drug has a favorable safety profile, hypersensitivity reactions to carboplatin are not rare.2–6 We showed that in women with an apparent allergic reaction to this drug and a positive skin test to carboplatin, readministration of the drug with a desensitization protocol is safe. Moreover, with this strategy, multiple treatments are feasible without allergic side effects.

Hypersensitivity is defined as an unexpected reaction with signs and symptoms inconsistent with the known toxicity of the drug. Previous reports on hypersensitivity reactions included both immediate and delayed reactions, as well as various nonspecific clinical presentations.7, 8, 10 A prerequisite for an immediate immune-mediated reaction includes symptoms compatible with an allergic reaction, signs appearing during or immediately after the administration of the drug, and documented preexposure to the offending drug. Indeed, all the adverse reactions in our study fulfilled these criteria. Thus, all the reactions coincided with administration of the drug and included the typical symptoms of an allergic reaction. Preexposure to carboplatin was evident in all patients before the adverse reaction to the drug. As previously described,5, 7, 8, 10 reactions to carboplatin tended to appear after a large number of treatment courses, usually after more than six treatment courses. Our group of patients fits into this observation, with a mean number of nine treatment courses before the first adverse reaction. Our observations support the suspected immunologic mechanism underling carboplatin hypersensitivity in these patients. This is supported, in addition, by the finding that all patients had a positive skin test to carboplatin. The skin test is an accepted diagnostic tool when an immediate allergic reaction is suspected.14 Carboplatin skin testing has been used previously for the diagnosis and prediction of hypersensitivity reactions to this drug.9, 15, 16 Markman et al.16 showed that a negative test substantially decreased the incidence of severe reactions. They also suggested that this test be used as a method of predicting the risk of developing an adverse reaction before carboplatin treatment. Although the positive predictive value of a positive skin test to carboplatin has not been defined, the combination of an immediate adverse reaction together with a positive skin test is highly indicative of an IgE-mediated reaction. Consequently, two operative strategies should be considered: avoiding the use of carboplatin in the future or using the desensitization protocol. The essential need for carboplatin in these patients and previous successful experience with different desensitization protocols4, 9–11 encouraged us to offer a desensitization protocol to our patients. Indeed, desensitization was well tolerated by most (95%) of our patients. These results are in agreement with previous publications, most of them including only case reports.3–5, 17 Two studies reported small groups of patients that were desensitized successfully. Markman et al.11 reported successful desensitization in four of five patients with either a hypersensitivity reaction or a positive skin test to carboplatin. All patients were premedicated extensively for 5 days before desensitization. Robinson et al.10 reported a high rate of success in desensitization to different chemotherapy agents. Eight patients with clinical carboplatin hypersensitivity were included in that study. Four of them underwent intravenous desensitization and tolerated further treatment courses with the drug. An allergic reaction was not confirmed by a skin test in that study. In contrast to these studies, Choi et al.18 recently reported their experience with desensitization in eight patients allergic to carboplatin. In 2 patients, a 6-hour desensitization protocol was attempted without success. Nevertheless, all patients tolerated uneventfully a longer infusion protocol.

The current study presents a large group of patients with a clinical allergic reaction to carboplatin confirmed by a positive skin test. The majority of the patients had moderate to severe reactions. The common attitude in such circumstances would have been to withhold carboplatin treatment in these patients. However, because carboplatin was an indispensable treatment, the majority of these patients were desensitized successfully and tolerated further treatments thereafter. The protocol we used is simple and short and does not require overnight hospitalization or additional premedication.

To our knowledge, the current study is the first large report of a successful desensitization protocol for a large sample of individuals with a proven allergic reaction to carboplatin. The data presented offer safe, convenient, and effective therapeutic strategy to patients who need this drug.

Ancillary