The paclitaxel/trastuzumab arm had improvements in response rate and time to progression but, in that subset, there wasn’t an overall survival advantage. You had to look at all 400-plus patients in the trial to see an improvement in survival.

For this patient, I would recommend paclitaxel, carboplatin and trastuzumab trial (Figure 5). We did a Phase III randomized trial, building upon the results of the pivotal trial to try to improve the outcome. In vitro data demonstrated that carboplatin is synergistic with either paclitaxel or docetaxel. Our trial addressed the very simple question: How does the addition of carboplatin affect outcomes?

We utilized the same schema used in the pivotal trial — paclitaxel every three weeks at 175 mg/m2 and carboplatin with an AUC of six. Our carboplatin dose was based on some of Edith’s work evaluating paclitaxel/ carboplatin every three weeks. We were very pleased that this laboratory work was carried through to the bedside.

In our trial of patients with HER2-positive (IHC 3+) metastatic disease, adding carboplatin to paclitaxel improved response rates to about 57% and doubled the time to progression to 14 months. We also saw an improvement of survival by nine months, although it’s not statistically significant. I don’t expect to see statistically significant improvement in survival — just as it wasn’t seen in the paclitaxel arm of the original pivotal trial — because we only had 191 patients. It’s also interesting to note that it took less than two months to see a response. This is a fast-acting regimen, considering that we don’t look for measurable disease more than once every two months.

Another important factor is that trastuzumab, like endocrine therapy, may be continued after chemotherapy. The pivotal trial mandated six cycles of chemotherapy, and we designed our trial in very much the same way. Three-quarters of the responding patients stopped chemotherapy after six cycles. In this strategy, chemotherapy is given in a synergistic combination with trastuzumab to induce a remission — in the parlance of leukemia — and after you have achieved the maximum response, then trastuzumab can be continued as maintenance therapy.

Dr Love: What was the tolerability of this combination regimen?

Dr Robert: Generally, this regimen was well- tolerated. We were concerned about myelosuppression from adding carboplatin to paclitaxel, and we did see more neutropenia and thrombocytopenia but no neutropenic fever. There was not an increased incidence of neuropathy. We saw slightly more fatigue, but we did not control for treatment with erythropoietin, so we cannot comment on whether or not this was related to anemia. It is important to note that this was not a dose-dense regimen, and we did not utilize prophylactic growth factors. If the counts were not adequate, then we would dose-reduce or dose-delay, and that partly explains the relatively favorable side-effect profile.

Dr Chittoor: When you give trastuzumab long-term, how do you monitor the cardiac status? I’ve had three patients who didn’t go into failure but had significant decreases in ejection fraction (EF). I checked EFs every six months. I realize that it is touted to be a reversible left ventricular dysfunction. Nonetheless, let us presume you do stop it and the EFs improve. Can you restart trastuzumab again?

Dr Perez: That’s a great question. When trastuzumab first became available, the package insert indicated that you should monitor left ventricular ejection fraction. We were routinely doing those evaluations, but we have completely abandoned that in our practice because we haven’t seen any correlation between asymptomatic decreases of ejection fraction and the development of congestive heart failure. Now the patients are managed on the basis of their physical symptoms.

In terms of clinical trials, we are doing very intense monitoring not only of left ventricular ejection fraction, but we are also looking at potential serum or plasma markers that may be predictive. Those studies are still in the investigational stages.

One of the challenges we have is that there is really not a database telling us that there’s a good correlation even between MUGAs and echos when we look at ejection fractions. There’s a vacuum in knowledge, and that’s one of the reasons we’ve abandoned routine evaluation of left ventricular ejection fraction.

It appears that in the few patients who have had endomyocardial biopsies after the development of trastuzumab-induced congestive heart failure, there is no documented anatomical vacuolization. This is very different than what is seen in anthracycline-induced cardiomyopathy, in which you see the vacuoles and can look at the damage. With trastuzumab, there’s no damage that is readily seen under the microscope. One of the theories is that this drug may be associated with stunning the myocardium, and many patients improve after decreases in ejection fraction.

One thing that we’re learning, just recently, is that this can also happen with chemotherapy. We have become very comfortable using anthracyclines, and we usually do not check ejection fractions. At the ASCO meeting this year, there was a wealth of information regarding this issue. We give AC x 4, the ejection fraction drops; we follow ejection fraction, and a few months later that number improves.

I think we have been sensitized to the issue of trastuzumab. We’ve been seeing a lot of ejection fractions, and we don’t really know how much that translates into the clinical symptomatology of our patients.

Dr Robert: Increasingly, I’ve heard that people are becoming more comfortable dropping the ventricular ejection fraction evaluation. The argument has been, if trouble arises, it will be early, within the first few months. That’s a good reason to initiate that test in the first few months. Afterwards, it doesn’t seem to be a problem — unlike with the use of doxorubicin. What I haven’t heard is evidence that the hypothesis is true — whether two or three years later, cardiac functioning is okay. Now, there aren’t that many patients who stay on trastuzumab or any regimen for years, but there are a few. If I’m giving a patient a drug that may affect their heart in a deleterious way, I should share those concerns with the patient. For patients who are responding to trastuzumab but develop congestive heart failure, you can continue trastuzumab and treat their cardiac condition with diuretics and ACE inhibitors and maintain a positive quality of life and good cardiac function.

Dr Love: Any comments from the group about the new data that Dr Robert was discussing with regard to carboplatin, docetaxel and trastuzumab?

Dr Firstenberg: We’ve been a big proponent of using platinums — both cisplatin and carboplatin — in patients with HER2-positive tumors, and then we use trastuzumab and vinorelbine.

Dr Love: Edith, what went into the evolution of using carboplatin versus cisplatin in some of these studies?

Dr Perez: We selected carboplatin based on the tolerability compared to cisplatin. A lower dosage of cisplatin needs to be studied, but we don’t have as much data on that drug, compared to carboplatin.

Dr Robert: The BCIRG has Phase II experience, in which cisplatin and carboplatin were used in combination with trastuzumab, and the response rates are comparable. The patient populations are not the same. (Table 6) Patients who received carboplatin had prior therapy, but the time to progression was longer in the carboplatin group. It’s a little bit like comparing “apples and oranges,” but I think all of us are convinced that carboplatin is better-tolerated than cisplatin.

Dr Robert: Another question for which we do not have an answer is how long should we continue trastuzumab after first-line therapy? For example, a patient receives carboplatin and paclitaxel, progresses, then you give them vinorelbine and the tumor progresses. Does it make sense to continue evaluating drugs that are either synergistic or additive? Is it time to stop?

We don’t have any evidence to direct us. One trial attempted to take patients who failed trastuzumab/taxane therapy and randomize them to vinorelbine versus vinorelbine and trastuzumab. That trial had poor accrual and it’s closed. It may resurface as an Intergroup trial.

Another issue is the schedule of trastuzumab. We have pharmacokinetic data from Brian Leyland-Jones that supports giving trastuzumab every three weeks instead of weekly, so our patients do not have to come into the office every week. This was a proof of principle trial, so we didn’t become involved with what was the best schedule for administering paclitaxel or carboplatin. We just wanted to demonstrate that adding carboplatin made a difference. Clearly, there are scheduling questions, and Dr Perez can tell us about her experience using weekly paclitaxel/carboplatin and trastuzumab, as opposed to the every-three-week schedule.

Dr Perez: At the NCCTG we have been interested in the taxane/carboplatin and the paclitaxel/carboplatin/trastuzumab combinations for a few years. We are completing a randomized Phase II trial — NCCTG-983252 — in which the basic question was the scheduling of the three drugs (Figure 6).

Dr Robert and colleagues undertook a study that would answer the question of how much carboplatin should be added to paclitaxel/trastuzumab, so we elected to look at paclitaxel/carboplatin/trastuzumab every-three-weeks versus given on a weekly basis.

In the every-three-week regimen, patients received paclitaxel 200 mg/m2 and carboplatin with AUC of six. The chemotherapy in the weekly regimen consisted of paclitaxel 80 mg/m2 with carboplatin AUC of two. The combination was given three out of four weeks. During the fourth week, patients received trastuzumab alone.

Our target accrual is 92 patients and enrollment is nearly completed.

We submitted an abstract to ASCO, documenting a significant difference in tolerability between the two regimens. Although the every-three-week regimen was fairly well-tolerated, we saw myelosuppression, some cases of febrile neutropenia, and some neuropathy.

However, when we looked at the tolerability of the weekly regimen, it was remarkably different in that we essentially did not see any myelosuppression or febrile neutropenia. And the difference in myelosuppression was not only in the white blood cell count but also with anemia. We also saw very few cases of neuropathy. Our conclusion was that we would recommend the weekly regimen be utilized in view of tolerability (Table 7).

The efficacy data on the first 70 patients is quite favorable. This is a randomized Phase II study, so we have to be a little bit careful. This is not really a randomized Phase III comparison of one regimen versus another, but it is a multi-institutional trial in which many of the patients were enrolled in community settings, so the patients actually were not enrolled at the Mayo Clinic.

Although the 95 percent confidence intervals of response rate overlapped a little bit, it’s actually much better to give the combination on a weekly schedule rather than every three weeks. We are currently using the weekly regimen in patients, even outside this trial. Based on the data from NCCTG-983252, weekly paclitaxel/carboplatin/trastuzumab has a better therapeutic ratio than using the drugs once every three weeks.

Case follow-up:
•	Patient treated with trastuzumab/carboplatin and docetaxel (75 mg/m2)
•	Very rapid clinical improvement; breast became less swollen and painful
•	After 3 cycles, no palpable abnormality in breast, ultrasound with marked improvement, pulmonary nodules decreased in size
•	Plan is for patient to complete 6 cycles of chemotherapy, undergo breast irradiation, and continue trastuzumab

Select publications

Home ·