While doxorubicin (Adriamycin) is among the most active single agents in the treatment of breast cancer and other solid tumors, its concomitant toxicity limits its use. Quality-of-life issues have driven the search for gentler,
ABSTRACT: While doxorubicin (Adriamycin) is among the most active single agents in the treatment of breast cancer and other solid tumors, its concomitant toxicity limits its use. Quality-of-life issues have driven the search for gentler, palliative treatments, especially for use in the frail and the elderly populations. Pegylated liposomal doxorubicin (PEG-LD) (Doxil) is being tested in various combinations to develop such a treatment regimen. Antitumor activity in mice was seen when PEG-LD was combined with the anti-EGFr monoclonal antibody C225, supporting its use in combination with biologicals. Clinical studies are also underway that combine PEG-LD with either vinorelbine (Navelbine), gemcitabine (Gemzar), or paclitaxel (Taxol), which are chemotherapeutic agents with non-overlapping toxicities compared with PEG-LD. In addition, a pegylated liposomal form of cisplatin (Platinol) is being tested in animal tumor models. Early evidence suggests that dosing regimens can be optimized so that such combinations provide antitumor activity comparable with conventional combination regimens but with acceptable toxicity. [ONCOLOGY 11(Suppl 11):63-68, 1997]
Doxorubicin (Adriamycin) is among the most active single agents used in the treatment of advanced breast cancer and other solid malignancies. However, the therapeutic index of this drug is often limited by its toxicities. These include nausea and vomiting and subacute toxicities such as alopecia, mucositis, and bone marrow suppression. Neutropenia is frequently encountered with doxorubicin at the dose intensities needed to achieve a meaningful response rate. Severe neutropenia can lead to febrile episodes and occasionally to septic infections that can be life-threatening. Moreover, the potential for irreversible cardiotoxicity limits cumulative exposure of doxorubicin to doses of less than 450 to 550 mg/m2.
Treatment strategies for advanced cancers vary geographically and by treatment centers. Depending on the disease setting, the goal of systemic chemotherapy can range from palliation to an intent to cure. Many medical oncologists believe that improved survival is not a realistic objective for systemic chemotherapy in patients who present with a high disease burden and/or multiple visceral metastatic sites. Clinical data support the antitumor activity of agents such as doxorubicin. However, some argue that the benefit of modestly improved survival among a small number of patients does not outweigh the risk of toxicity to all patients who receive such therapy, particularly elderly and/or frail patients who are less likely to tolerate therapy.
In some treatment centers, therefore, patients with advanced cancers are not treated with chemotherapy, but are given supportive care to palliate the signs and symptoms of the disease. This debate has raised awareness of the importance of patients quality of life and has driven a search for treatment options that provide responses in a significant number of patients, and/or delay disease progression for a meaningful period of time with minimal toxicity.
Preclinical studies of pegylated liposomal doxorubicin (PEG-LD) (Doxil) and experience in the treatment of Kaposis sarcoma (KS) suggest that compared with standard doxorubicin, pegylated liposomes (Stealth) deliver a greater proportion of an injected dose of doxorubicin to tumor sites. If this were also the case in other human cancers, one might reasonably expect PEG-LD to have similar antitumor activity to that of doxorubicin, but at a lower dose intensity, which would produce less severe toxicity.
This expectation provides a rationale, for example, for developing PEG-LD as single-agent therapy for advanced breast cancer among elderly patients. The goal would be a tumor response rate comparable with doxorubicin but with a dose and schedule of PEG-LD that minimizes the frequency and severity of nausea, vomiting, neutropenia, mucositis, alopecia, and cardiotoxicity.
This reasoning may extend to combinations of PEG-LD with several noncytotoxic biologic agents, which are the subject of ongoing clinical investigations. In general, such agents produce minimal toxicity but are not sufficiently active on their own.
Humanized monoclonal antibodies directed against receptors overexpressed on malignant cells such as p185HER2 or the EGFr, appear to fall into this category.[2,3] Peptide and protein inhibitors of angiogenesis may also need to be combined with a cytotoxic agent for meaningful antitumor activity. By combining these agents with PEG-LD, the good tolerance to each may be maintained while additive or perhaps synergistic antitumor activity may be achieved.
This theory has been tested preclinically using a combination of the anti-EGFr monoclonal antibody C225 and PEG-LD against a xenograft of the A431 tumor implanted in nude mice. As shown in Figure 1, a combination of PEG-LD and the C225 antibody produced an antitumor effect superior to either of these agents alone, to doxorubicin alone, and to a combination of the antibody plus doxorubicin. These results support the therapeutic rationale of combining PEG-LD with biologics.
Two principles guide the selection of cytotoxic agents for use in combination regimens: (1) they should be individually active in the selected tumor type and (2) have nonoverlapping dose-limiting toxicities. PEG-LD has been shown to be active as a single agent in refractory ovarian cancer and as primary therapy for recurrent breast cancer.[1,5] While generally manageable in both of these settings, epithelial cell toxicity manifesting itself as palmar-plantar erythrodysesthesia (PPE, hand-foot syndrome) may limit the amount of PEG-LD patients are able to tolerate. Therefore, it is reasonable to explore combinations of PEG-LD with other agents proven to be active in these tumor types but with toxicities other than skin.
PEG-LD Plus Vinorelbine
The dose-limiting toxicity of vinorelbine tartrate (Navelbine) is granulocytopenia. In combination with doxorubicin, vinorelbine produced a 57% overall objective response rate as first-line therapy for advanced breast cancer. However, the incidence of grade 4 granulocytopenia was 83%, with 8% requiring hospitalization due to febrile neutropenia and one septic death. Substitution of doxorubicin with PEG-LD in this combination is being explored as a means of maintaining the favorable tumor response of the combination while reducing the incidence of hematological toxicity. Vinorelbine would not be expected to contribute to the skin toxicity seen with PEG-LD.
PEG-LD Plus Gemcitabine
Gemcitabine (Gemzar) given alone has been reported to have meaningful activity in non-small-cell lung, breast, ovarian, and head-and-neck cancers and a favorable safety profile. Only modest levels of traditional cytotoxicities such as myelosuppression, nausea and vomiting, and alopecia are seen with gemcitabine administered on a weekly schedule at doses less than 1,000 mg/m2. Although maculopapular skin rashes have been reported secondary to injection of gemcitabine, particularly in Asian patients, the timing of the appearance of the rash (within 1 to 2 days of injection) and the anatomical sites (chest, abdominal wall, upper arms and thighs) suggest that the underlying cause of this toxicity will not aggravate the palmar-plantar erythrodysesthesia seen with PEG-LD. Given the non-overlapping toxicity profiles of gemcitabine and PEG-LD and the complimentary single-agent activity of the two in breast and ovarian cancer, combinations of these agents are being tested in ongoing pilot phase I dose-finding trials.
PEG-LD Plus Paclitaxel
The excitement generated by Gianni et al who reported a greater than 90% objective response rate in metastatic breast cancer for a paclitaxel/doxorubicin combination is tempered by the rather unfavorable side effect profile of this regimen. Severe, febrile neutropenia was common and peripheral neuropathy occurred in one third of the patients. Perhaps more troubling was the development of congestive heart failure (CHF), albeit reversible in 18% of women after a median of 480 mg/m2 doxorubicin. These results raise the specter of paclitaxel-related enhancement of doxorubicin cardiotoxicity.
Studies of single-agent doxorubicin have demonstrated that CHF is very uncommon at cumulative doses of less than 500 mg/m2. Dosing schedules of doxorubicin designed to lower peak plasma concentration are known to reduce cardiotoxicity, suggesting that the peak dose (Cmax) in plasma after a typical doxorubicin infusion contributes to development of this toxicity. Peak levels of bioavailable doxorubicin are substantially reduced following PEG-LD administration, and preclinical results indicate that PEG-LD causes fewer morphological changes in heart tissues relative to free doxorubicin in three species: rat, dog, and rabbit.
These encouraging preclinical findings are supported by results of a pilot clinical study by Berry et al. These authors performed endomyocardial biopsies on a series of AIDS-KS patients who had received cumulative doses of PEG-LD ranging from 469 to 860 mg/m2. This group was compared with two historical control groups treated with doxorubicin. One group was matched with respect to cumulative exposure to doxorubicin and the other with respect to both cumulative dose and administered dose. Biopsy specimens were examined microscopically and scored according to the 0 (no evidence of anthracycline-specific damage) to 3.0 (severe diffuse myocyte damage; greater than 35% of all cells) scale introduced by Billingham and colleagues. The mean biopsy score for the PEG-LD group was 0.5 (± 0.6). For the two control groups the mean biopsy scores were 2.1 (± 0.7) and 1.4 (± 0.65), respectively. The comparisons between the mean PEG-LD scores and the control groups were highly significant in both cases (P < .001).
These finding support the rationale for combining paclitaxel with PEG-LD. Both drugs have demonstrated activity in breast and ovarian cancer. With respect to toxicities, the incidence of severe neutropenia and peripheral neuropathy are lower for PEG-LD than for paclitaxel. Also paclitaxel causes relatively little skin toxicity. In addition, preclinical and early clinical biopsy results strongly suggest that PEG-LD produces less damage to the myocardium relative to comparable cumulative doses of doxorubicin. Thus the cardioprotective effect of PEG-LD may translate into a reduced risk of cardiotoxicity relative to the highly active paclitaxel-doxorubicin combination. Based on these considerations, several phase I dose-finding trials of PEG-LD and paclitaxel have been launched.
The favorable pharmacokinetic and tissue distribution patterns of PEG-LD are a consequence of the liposome carrier, not the drug. The encapsulated drug becomes bioavailable only after the liposome itself enters tissues, including tumors. To the extent that other drugs can be successfully formulated in the same liposomes used for PEG-LD, they may have the same benefit. Indeed, several antitumor drugs have been reported to show improved tolerance and comparable or better tumor response when formulated in a PEG-LD-like Stealth liposome.
Stealth Cisplatin (SPI-77)
Cisplatin (Platinol) is active alone and in combination chemotherapy regimens against a wide range of epithelial malignancies including testicular, ovarian, head and neck, lung, bladder, and cervical cancers. Cisplatin chemotherapy is often limited by side effects that prohibit continued treatment. In addition, some tumors are initially resistant or acquire cisplatin resistance with continued exposure. The major dose-limiting, cisplatin-induced toxicity in humans is renal toxicity, although significant nausea and vomiting, ototoxicity, peripheral neuropathy, and myelotoxicity are also induced by cisplatin administration. Attempts to ameliorate cisplatin-induced toxicity and/or resistance have focused on the development of platinum derivatives that are less toxic and/or more active than the parent compound.[14-18] Alternative approaches include altering the pharmacology of the drug by modifying the treatment schedule, hydrating patients prior to and during therapy, administering renal protectant therapy, or encapsulating the drug within liposomes. [19,20]
SPI-77 is a formulation of cisplatin encapsulated in virtually the same type of liposome as PEG-LD. SPI-77 exhibits plasma pharmacokinetics characteristic of sterically stabilized (Stealth) liposomes. These are long circulation, high Cmax and area-under-the-plasma concentration vs time curve (AUC), and low clearance and volume of distribution compared with nonliposomal cisplatin (Figure 2). In vitro leakage studies suggest that plasma levels of platinum primarily or solely represent liposomal cisplatin, (ie, drug that is in liposomes and not free or bound to proteins).
The therapeutic activity of SPI-77 has been evaluated and compared with nonliposomal cisplatin in various tumor models, including the C26 colon carcinoma in Balb/c mice and a xenograft of the NCI-H82 small-cell lung tumor in athymic mice (Figure 3). While SPI-77 showed meaningful antitumor activity in both of these tumor models, cisplatin was only effective in the NCI-H82 xenograft model; carboplatin (Paraplatin) was ineffective in both. Although SPI-77 only occasionally produced complete tumor responses, it did cause a persistent inhibition of tumor growth during and after treatment. In many animals, tumors grew slowly to an intermediate size and then were apparently arrested, with little additional growth evident. Although cisplatin treatment resulted in better inhibition of tumor growth in both trials using the NCI-H82 xenograft models, SPI-77 was more effective in producing a prolonged response to treatment, with persistent inhibition of tumor growth.
In preclinical safety studies in support of the SPI-77 IND submission, the primary adverse effects induced by single- and multiple-dose treatments with SPI-77 in rodent species are reported to be hepatic and biliary tract toxicity. Hepatic toxicity was seen only at high doses in non-rodent species. The dose relationship of these toxicities seen in the rabbit support the notion that the intensified hepatic/biliary toxicity of SPI-77 in small rodents may be a function of the saturation of the mononuclear phagocyte system (MPS) by the lipid component of the formulation. Smaller species may have less MPS capacity and, therefore, be more sensitive to the potential decrease in macrophage function.
In dogs given multiple doses of 150 or 300 mg/m2 SPI-77, there was no evidence of renal, hepatic, or nervous system toxicity and no significant gastrointestinal toxicity. The only notable findings were minimal cyclical and reversible decreases in erythrocyte-related hematology values. In contrast, treatment of dogs with cisplatin at a dose of 70 mg/m2 every third week to a cumulative dose of just 280 mg/m2 (four treatments), causes significant renal toxicity, and deafness and blindness in 20% of animals. Lipoidal glomerulopathy, apparently reversible, was seen in rats, but not in rabbits, dogs, or monkeys. All changes showed a clear dose-response relationship within species, and, accounting for rather marked differences in species sensitivity, were similar among all five species tested.
In a small pilot study, 3 female monkeys received either 3, 10, or 30 mg/kg (36, 120, or 360 mg/m2) SPI-77 every third week for a total of three treatments. The only notable findings were minimal to mild, dose-related decreases in red blood cell count, hematocrit, and hemoglobin. Minimal to mild transient elevations of AST and ALT were seen in all three animals after the first and second doses, but were only seen in the high-dose monkey after the third treatment. Cholesterol levels increased dose-dependently, peaked 24 hours after treatment, and returned to normal range prior to the next dose. No clinical chemistry changes suggestive of renal toxicity were seen, and there were no changes in serum electrolyte or urinalysis parameters.
No evidence was found to suggest that the mechanism of toxicity of SPI-77 was different among different species. Renal toxicity was notably absent in all species tested (rat, mouse, rabbit, dog, and monkey). Despite the presence of significant multiple-dose hepatic toxicity in rodents, SPI-77 given as single and repeated treatments in three non-rodent species (rabbits, dogs, and monkeys) at very high doses, produced relatively little toxicity. This suggested that SPI-77 has a margin of safety acceptable to support clinical development. Based on these favorable safety results and evidence of antitumor activity comparable or better than cisplatin in animal models, phase I dose-finding studies of SPI-77 are underway.
Vincristine (Oncovin) is used clinically both as a single agent and in combination regimens for the treatment of hematologic malignancies, head and neck cancer, Kaposis sarcoma and lung cancer. Early work with conventional liposome-encapsulated vincristine showed no improvement in safety or therapeutic activity relative to the free drug. However, the improvement in antitumor activity seen with long circulating Stealth formulations of PEG-LD and SPI-77 led to the investigation of a Stealth liposome formulation of vincristine.
Pharmacokinetic studies showed that Stealth liposome encapsulated vincristine (S-Vinc) prolonged the drugs distribution phase plasma half-life in rats from 0.22 to 10.5 hours. While there was no significant difference in LD 50 between encapsulated and free drug (at doses of 2.5 mg/kg, given intravenously), mice given sublethal doses of S-Vinc experienced significantly less weight loss compared with animals receiving the same dose of free vincristine. Compared to free drug, S-Vinc was more active against intraperitoneally and subcutaneously implanted tumors. In a subcutaneously-implanted murine colon tumor model, multiple doses of free drug did little to retard tumor growth, but S-Vinc slowed tumor growth and improved long-term survival with several dosing regimens (Figure 4).
Combinations of PEG-LD with other cytotoxic agents with non-overlapping toxicity profiles are being explored clinically, including vinorelbine, gemcitabine, and taxanes. Early evidence suggests that dosing regimens can be optimized so that such combinations provide antitumor activity comparable with conventional combinations but with acceptable toxicity.
The rationale for combining PEG-LD with biologicals, such as humanized antibodies directed against receptors overexpressed on tumor cells, is supported by positive preclinical findings of synergistic antitumor activity of a PEG-LD-antiEGFr antibody regimen. Such combinations may boost the therapeutic utility of PEG-LD while maintaining its favorable safety profile. Elderly and/or frail patients unable to tolerate aggressive cytotoxic chemotherapy may benefit from this approach.
Stealth liposomal delivery of antitumor agents other than doxorubicin appears feasible. Formulations of vincristine and cisplatin have been shown to have superior anti-tumor activity in animal models with equal or better tolerance. Phase I studies of SPI-77, a Stealth liposome formulation of cisplatin, are underway.
1. Ranson MR, Carmichael J, OByrne K, et al: Treatment of advanced breast cancer with Stealth liposomal doxorubicin (CAELYX): Results of a multicenter phase II trial. J Clin Oncol 15:3185-3191, 1997.
2. Baselga J, Tripathy D, Mendelsohn J, et al: Phase II study of weekly intravenous recombinant humanized anti-p185 HER2 monoclonal antibody in patients with HER2/neu-overexpressing metastatic breast cancer. J Clin Oncol 14:737-744, 1996.
3. Chrysogelos SA, Dickson RB: EGF receptor expression, regulation and function in breast cancer. Breast Cancer Res Treat 29:29-40, 1994.
4. Folkman J: Fighting cancer by attacking its blood supply. Sci Am 275:150-154, 1996.
5. Muggia FM, Hainsworth JD, Jeffers S, et al: Phase II study of PEG-LD in refractory ovarian cancer: Antitumor activity and toxicity modification by liposomal encapsulation. J Clin Oncol 15:987-993, 1997.
6. Hochster HS: Combined doxorubicin/vinorelbine (Navelbine) therapy in the treatment of advanced breast cancer (suppl 5). Semin Oncol 22(2):55-59, 1995.
7. Gianni L, Munzone E, Capri G, et al: Paclitaxel by 3-hour infusion in combination with bolus doxorubicin in women with untreated metastatic breast cancer: High antitumor efficacy and cardiac in a dose-finding and sequence-finding study. J Clin Oncol 13:2688-2699, 1995.
8. Von Hoff DD, Layard MW, Basa P, et al: Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med 91:710-717, 1979.
9. Legha SS, Benjamin RS, Mackay B, et al: Reduction of doxorubicin cardiotoxicity by prolonged continuous intravenous infusion. Ann Intern Med 96:133-139, 1982.
10. Gabizon A, Catane R, Uziely B, et al: Prolonged circulation time and enhanced accumulation in malignant exudates of doxorubicin encapsulated in polyethylene glycol coated liposomes. Cancer Res 54:987-992, 1994.
11. Berry G, Billingham M, Alderman E, et al: Reduced cardiotoxicity of PEG-LD (pegylated liposomal doxorubicin) in AIDS Kaposis sarcoma patients compared to a matched control groups of cancer patients given doxorubicin. Proc Am Soc Clin Oncol 15:843, 1996.
12. Billingham ME, Mason JW, Bristow MR, et al: Anthracycline cardiomyopathy by morphologic changes. Cancer Treat Rep 62:865-872, 1978.
13. Loeher PJ, Einhorn LH: Drugs five years later: Cisplatin. Ann Intern Med 100:704-713, 1984.
14. Lelieveld P, van der Vijgh WJF, Veldhuizen RW, et al: Preclinical studies on toxicity, antitumor activity and pharmacokinetics of cisplatin and three recently developed derivatives. Eur J Cancer Clin Oncol 20:1087-1104, 1984.
15. Mathe G, Kidani Y, Segiguchi M, et al: Oxalato-platinum or L-OHP, a third-generation platinum complex: An experimental and clinical appraisal and preliminary comparison with cis-Platinum and carboplatinum. Biomed Pharmacother 43:237-250, 1989.
16. Schilder RJ, LaCreta FP, Perez RP, et al: Phase I and pharmacokinetic study of ormaplatin (tetraplatin, NSC 363812) administered on a day 1 and 8 schedule. Cancer Res 54:709-717, 1994.
17. McKeage MJ, Mistry P, Ward J, et al: A phase I and pharmacology study of an oral platinum complex, JM216: Dose-dependent pharmacokinetics with single-dose administration. Cancer Chemother Pharmacol 36:451-458, 1995.
18. Kelland RL, McKeage MJ: New platinum agents. A comparison in ovarian cancer. Drugs Aging 5:85-95, 1994.
19. Steerenberg PA, Storm G, de Groot G, et al: Liposomes as a drug carrier system for cis-diamminedichloroplatinum (II). I. Binding capacity, stability and tumor growth inhibition in vitro. Int J Pharmaceutics 40:51-62, 1987.
20. Potkul RK, Gondal J, Bitterman P, et al: Toxicities in rats with free versus liposomal encapsulated cisplatin. Am J Obstet Gynecol 164:652-658, 1991.
21. Ogilvie GK, Fettman MJ, Jameson VJ, et al: Evaluation of a one-hour saline diuresis protocol for administration of cisplatin to dogs. Am J Vet Res 53:1666-1669, 1992.
22. Layton D, Trouet A: A comparison of the therapeutic effects of free and liposomally encapsulated vincristine in leukemia mice. Eur J Cancer 16:949-950, 1980.
23. Allen TM, Newman MS, Woodle MC, et al: Pharmacokinetics and anti-tumor activity of vincristine encapsulated in sterically stabilized liposomes. Int J Cancer 62:199-204, 1995.