Is There a Role for Intraperitoneal Chemotherapy in the Management of Ovarian Cancer?

Since the 1970s, dose intensity hasbeen the siren song of the medicaloncologist. Simply stated, the words of the song keep repeating: "More isbetter." The tendency of ovarian cancer is to spread via intraperitonealdissemination and to remain grossly confined to the peritoneal cavity throughoutmuch of its natural history. Fortunately or unfortunately, this tendency enablesan alternative to stem cell-supported high-dose therapy, by which we canachieve even greater dose intensity.

Proof of Efficacy ofIntraperitoneal Therapy

Gynecologic and medical oncologists have studied intraperitonealchemotherapy for 22 years. Several observations over that time form the basis ofintraperitoneal therapy. First, it is feasible to administer the most activeagents for ovarian carcinoma directly into the peritoneal cavity. Second, interms of peak drug levels, as well as area under the concentration-time curve(AUC), the intraperitoneal route offers significant pharmacokinetic advantagesover intravenous (IV) administration. Third, significant systemic exposure tocisplatin (Platinol) is still achieved because the drug is taken up into thevascular compartment to a great extent. Fourth, phase I and II studies havereported numerous responses.

While these rationales were being drawn, however, severalpotential problems were largely downplayed: Extraperitoneal spread was notinfrequent; drugwas unevenly distributed in the peritoneal cavity because ofadhesions; intraperitoneal drug had poor depth of penetration into larger (>0.5 cm) nodules; and there was a lack of evidence of the value of doubling doseintensity instead of using the usual dose ranges of agents. These problems,along with the adverse effects that are unique to intraperitoneal therapy, aswell as the more complex logistics of intraperitoneal drug administration,mandate that there be a clinically documented advantage to the intraperitonealroute in terms of efficacy before this approach is adopted as part of standardcare.

As Dr. Markman correctly points out, the introduction of a morecomplex and potentially more toxic approach requires a positive randomized phaseIII trial. He cites three different clinical settings in which he believesintraperitoneal therapy would be a rational option for ovarian carcinoma (Table4 of the Markman article). Many of the reports on which this is based, however,do not distinguish between patients who responded to front-line therapy (andhence can be expected to be chemosensitive) and those who failed to respond tofront-line therapy.

Intraperitoneal Regimens as Second-Line Therapy

In a 1991 report, Dr. Markman and colleagues detailed studies of72 patients treated with intraperitoneal platinum-based regimens afterfront-line therapy.[1] In those who responded to front-line therapy, a highresponse rate was observed (59%), whereas those who failed to respond tofront-line platinum exhibited only a 9% response rate. Even in patients who werechemosensitive, IV platinum-based second-line therapy produced similar results.This raises serious questions about any defined role for intraperitonealchemotherapy in the second-line setting. Most telling, there are no randomizedphase III trials to establish any advantage for intraperitoneal therapy.

Intraperitoneal Therapy in the Front-Line Setting

In contrast to the second-line setting, three randomized phaseIII trials have compared IV and intraperitoneal regimens for front-linetherapy.[2-3; D. Armstrong, personal communication, 2000.] In his article, Dr.Markman discusses these extensively. The first of these, a Southwest OncologyGroup (SWOG)/ Gynecologic Oncology Group (GOG) trial[2] compared IVcyclophosphamide (Cytoxan, Neosar) plus either IV or intraperitoneal cisplatinin patients with disease up to 2 cm in diameter. Results showed a statisticallysignificant survival advantage for the intraperitoneal regimen. Unfortunately,however, this study suffers from several problems.

The initial goal of the study was to accrue 400 patients to seeka hazard ratio of 0.67. At the midpoint of accrual, the investigators decided toextend accrual to 650 patients to accrue sufficient patients with < 0.5-cmdisease to allow a subset analysis on the basis that only these patients couldbe expected to benefit. The final analysis included all 650 patients anddetected a significant hazard ratio of 0.77. The original goal of the study—ahazard ratio of 0.67— was never achieved. Furthermore, the subset of patientswith < 0.5-cm residual disease did not significantly benefit fromintraperitoneal therapy (median survival: 51 vs 46 months). These problems raiseserious questions about any conclusion that the intraperitoneal regimen wassuperior.

The second study, a GOG-SWOG study[3] compared IV paclitaxel(Taxol)/cisplatin to a dose-intense regimen of two cycles of carboplatin(Paraplatin), AUC 9, followed by IV paclitaxel plus intraperitoneal cisplatin.This trial was never intended as a study of intraperitoneal vs IV therapy.Instead, the study compared a standard regimen to a dose-intense regimenpatterned after an approach developed at Memorial Sloan-Kettering Cancer Center.This study showed a significant improvement in progression-free survival; butthe difference in survival is only of borderline significance. Thedose-intensive regimen was significantly more toxic.

Thus, while it is true that two randomized phase III trials haveshown a possible advantage for an intraperitoneal regimen, both studies havesufficient problems in that neither can serve as the basis for usingintraperitoneal therapy as part of standard care. Dr. Markman perhaps stated itbest: "Despite the observed toxicity, the results of this study continuedto stimulate interest with regard to defining a role for intraperitoneal drugdelivery in ovarian cancer." In short, even Dr. Markman feels there is nodefined role for intraperitoneal therapy at this point.

The third trial, a GOG study[4] compared IV paclitaxel/cisplatinto IV paclitaxel followed by intraperitoneal cisplatin and paclitaxel. Thisstudy—which has completed accrual and is now maturing—should answer once andfor all whether intraperitoneal therapy plays any front-line role in ovariancancer. In the absence of results from this third effort, however,intraperitoneal regimens should not be used for ovarian carcinoma outside ofclinical trials.

If, after all of this, there remain skeptics who insist thatintraperitoneal therapy has a proven advantage, such individuals should askthemselves these questions: If the issue were settled, why was a third trialnecessary? If the advantages are clear, why do so few physicians dealing withovarian carcinoma recommend intraperitoneal therapy? More is better, more isbetter, more is better. . . . It is, isn’t it?

References:

1. Markman M, Reichman B, Hakes T, et al: Responses tosecond-line cisplatin-based intra- peritoneal therapy in ovarian cancer:Influence of a prior response to intravenous cisplatin. J Clin Oncol9:1801-1805, 1991.

2. Alberts D, Lui P, Hannigan E, et al: Intra-peritonealcisplain plus intravenous cyclophosphamide versus intravenous cisplatin plusintravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med335:1950-1955, 1996.

3. Markman M, Bundy B, Benda J, et al: Randomized phase 3 studyof intravenous cisplatin/paclitaxel versus moderately high-dose intravenouscarboplatin followed by intravenous paclitaxel and intraperitoneal cisplatin inoptimal residual ovarian cancer: An intergroup trial (GOG, SWOG, ECOG)(abstract). Proc Am Soc Clin Oncol 17:361a, 1998.