Gary Culliton reports on a recent meeting on ‘Advances in Ovarian Cancer’, where delegates got the latest updates on clinical trials into the causes and treatment of the disease.
Worldwide, there are 220,000 new cases of ovarian cancer and 140,000 deaths each year, it was revealed at the ‘Advances in Ovarian Cancer’ meeting in Dublin’s Mater Private Hospital recently.
Dr Robert Soslow, Gynaecologic Pathologist at New York’s Memorial Sloan-Kettering Cancer Center, made the case that ovarian cancer constituted a number of separate diseases. Among these were high-grade serous carcinoma, clear cell carcinoma, endometrioid carcinoma and low-grade serous carcinoma.
Many high-grade serous carcinomas probably arise in the fallopian tube, the delegates heard. BRCA1 mutation gives rise to serous carcinomas with characteristic morphological features. These features may facilitate recognition of tumours with a relatively better prognosis due to platinum sensitivity, anti-tumour immunity or ‘de-bulkability’. They also allow recognition of at-risk tissues in the index patient and in at-risk family members.
Ovarian cancer might be thought of as a disease of DNA instability. However, unlike in other cancers, there are no overwhelmingly evident driver mutations in ovarian cancer (Levine D, TCGA). A study limited to Stage III/IV BRCA-associated ovarian cancer found median survival of 7.5-to-eight years (Chetrit, JCO, 2008). Some of this improvement may be due to better responses to platinum-based chemotherapy, Dr Noah Kauff, of the Gynaecology and Clinical Genetics service at Memorial Sloan-Kettering, said.
Platinum sensitivity is associated with improved survival. However, even in the absence of platinum sensitivity, BRCA mutation status was associated with improved outcome (Gallagher DJ, Ann Oncol, 2011).
Poly ADP ribose polymerase (PARP) inhibitors are a major class of drugs in this disease. Patients who have the best response to PARP are also those patients who are platinum-sensitive. Among BRCA-mutated patients, a 200mg twice-daily dose of olaparib showed a 28 per cent response rate and a seven-month progression-free interval. With a dose of 400mg, people who had finished platinum-sensitive recurrence therapy with gemcitabine/carboplatin were examined (Lederman, ASCO, 2011). A total of 22 per cent of these carried the BRCA1 or BRCA2 mutation. Those patients with the mutation did better than those without it.
There was a near doubling of progression-free survival (PFS) from 4.8 months to 8.4 months in those patients who remained on olaparib.
Dr Paul Sabbatini, Attending Medical Oncologist at Memorial Sloan-Kettering, focuses his clinical practice on women with gynaecological malignancies.
His research focuses on the period of remission after initial treatment of ovarian cancer with chemotherapy. Dr Sabbatini looked at current standards for ovarian cancer treatment and primary therapy. He reviewed research presented at ASCO 2011 and offered a preview of how clinical treatment might change.
Dr Sabbatini looked at the remission population, proposing that the second remission population was one in which much could be learned about new therapies and where new agents that prolong remissions might be of value (Dizon et al, Journal of Clinical Oncology 2002).
The disease does not start in a systematic way and Dr Sabbatini said that he considers it is unlikely that early imaging studies will ever be sufficient. He expects that biomarkers such as YKL-40 and HE4 will be of interest in the unselected population.
In terms of primary treatment of ovarian cancer, the standard approach is maximum de-bulking surgery and then standard taxane and platinum-based chemotherapy. If the patient is optimally de-bulked (maximum diameter of residual less than 1cm), PFS of 24-to-28 months and overall survival of 52-to-65 months is possible. If suboptimal, overall survival consistently ranges around 38 months.
The most recent randomised trials suggested it was not helpful to return for more surgery after chemotherapy, where surgical de-bulking in the beginning could not be attempted because there was too much disease.
The Gynaecologic Oncology Group (GOG) 182-ICON5 was the first study to consider whether adding gemcitabine, liposomal doxorubicin or topotecan to primary therapy might secure any benefit. Mixing and matching standard chemotherapy at the beginning of a patient’s course of treatment is unlikely to change events or downstream endpoints, Dr Sabbatini said. This trial showed there was nothing better than carboplatin and paclitaxel in the upfront setting (Bookman, J Clinical Oncology, 2009).
Intraperitoneal (IP) therapy has to be considered in upfront treatment of ovarian cancer. Three studies support the notion that IP therapy has benefits in ovarian cancer. Alberts et al (NEJM, 1996) found there was an eight-month survival advantage with IV cyclophosphamide with either intravenous or intraperitoneal cisplatin. Maurie Markman (JCO, 2001) administered IV paclitaxel. A high dose of IV carboplatin was introduced to de-bulk people before intraperitoneal cisplatin treatment. There was a 10-month survival advantage.
Deborah Armstrong (NEJM, 2006) was more straightforward and used IV paclitaxel with IV cisplatin. This study looked at weekly paclitaxel too (IP paclitaxel was also administered on day eight). Toxicity was high. In the GOG 182-ICON5 trial, some 46 per cent of people had grade III/IV toxicity.
With a dose reduction of cisplatin from 100mg down to 75mg, the toxicity profile changed significantly.
The neuropathy decreased substantially.
Ovarian cancer has more vascular endothelial growth factor (VEGF) in its acidic form than any other tumour type that has been examined. Like many tumours, low VEGF in this disease predicts higher survival and high VEGF predicts lower survival. Bevacizumab has a single-agent response rate, which is quite atypical in ovarian cancer.
The major trial for platinum-sensitive patients reported at ASCO was the OCEANS study (Carol Aghajanian, ASCO 2011). It looked at gemcitabine and carboplatin, with or without bevacizumab. It considered a robust platinum-sensitive group.
The response rate was significantly higher in patients receiving bevacizumab (79 per cent versus 57 per cent). There was a four-month PFS advantage for giving bevacizumab and interim survival data for platinum-sensitive recurrent patients showed an advantage of 29 versus 35 months.
The study concluded that these drugs prolong PFS. Thrombolic events, central nervous system bleeding and neutropaenia were not significantly different.
The GOG218 study compared giving: A) paclitaxel, carboplatin and placebo; B) paclitaxel, carboplatin plus bevacizumab solely during chemotherapy; and C) paclitaxel, carboplatin and continued bevacizumab through chemotherapy and maintenance (for up to 22 cycles). About 30 per cent of patients were optimal and 40 per cent were sub-optimal. The rest were stage IV.
Bevacizumab rose to the top in the Gynaecologic Oncology Group study. The response rate was higher. A decision was made to add this agent to front-line therapy in a bid to improve outcome. The proportion of patients with PFS at six months was 40 per cent, clearly higher than anything else. The ICON7 trial produced similar data to this. There was an increase in median PFS of 10.3 months to 14.1 months — a 3.8-month advantage for continuing the bevacizumab for 22 cycles of therapy. The same benefit accrued if bevacizumab was given later.
Also looking at upfront therapy for ovarian cancer, the ICON7 trial included high-risk, early-stage patients, as well as Stage III and IV patients. The dose here was half that in the Gynaecologic Oncology Group study. The PFS in favour of the bevacizumab arm was 19.8 months, compared to 17.4 months (ASCO 2011). This represented a 2.8-month PFS advantage for continuing bevacizumab for 12 cycles.
Data presented at ESMO 2010 showed no survival advantage for bevacizumab. However, this was updated this year. An emerging survival difference was reported among a stratum of patients who were suboptimally de-bulked at the beginning or who had bulky disease when they started treatment. There was a survival difference among this high-risk group — 28 months versus 36 months. Thus, while there appeared to be no survival advantage in the low-risk group, there was a survival advantage in the high-risk group (patients with bulk disease when they started their therapy).
A pilot study used IP cisplatin and IP paclitaxel with standard IP bevacizumab (Konner, MSKCC 2007). Hypertension was 10 per cent lower. The GOG 0252 trial showed this combination could be given without risk of toxicity. The GOG 0252 study used weekly paclitaxel in two of the arms, versus the standard three-week paclitaxel schedule. Carboplatin was substituted for cisplatin. Bevacizumab was in all three arms of this trial.
A Japanese study (JGOG, Kalsumata, Lancet, 2009) looked at paclitaxel and carboplatin therapy every three weeks, versus weekly paclitaxel. A major progression-free advantage for weekly paclitaxel — 17 to 28 months — was observed. There was an overall survival advantage of 72 per cent versus 65 per cent.
GOG 0262 — which has not reported – is an important trial for upfront treatment. Weekly paclitaxel was compared to three-weekly paclitaxel. The bevacizumab treatment was made optional. Sorafenib had some activity, as did temsirolimus.
Dr Sabbatini then considered relapse therapy. Taxane therapy, etoposide, vinorelbine and altretamine are among the standard agents that have activity in relapsed ovarian cancer. Paclitaxel, topotecan, liposomal doxorubicin and gemcitabine have been compared, one to another, in Phase III trials. For single-agent therapy, there is a 20-30 per cent response rate if the patient is platinum sensitive. For patients who are platinum resistant, there is a 15-20 per cent response rate.
The only single-agent drug that was added to the list at ASCO this year was trabectedin, which had a 23 per cent platinum-sensitive response rate.
There are three randomised trials that lend themselves to advice about how to treat people with platinum-sensitive, recurrent ovarian cancer. The first one (ICON4/AGO-OVAR, Parmar M et al, Lancet 2003) compared paclitaxel and carboplatin to carboplatin alone. This showed both a PFS and an overall survival advantage. The AGO group looked at gemcitabine/carboplatin versus carboplatin. There was a three-month progression-free advantage for the doublet.
There were updated data about platinum-sensitive groups at ASCO. The CALYPSO trial, presented at ASCO last year, compared liposomal doxorubicin combined with carboplatin, versus carboplatin combined with paclitaxel. A PFS advantage of 11.3 months was demonstrated. At ASCO this year, it was revealed that overall survival was 31.5 months in both arms.
Moreover, there was no difference in PFS in a very-platinum-sensitive subgroup of the trial. “It’s possible to retreat people with platinum-sensitive disease, particularly with either liposomal doxorubicin/carboplatin or with gemcitabine/carboplatin,” said Dr Sabbatini. “These patients can be spared additional alopecia and neuropathy. Paclitaxel can be used later.”
As patients are retreated with carboplatin, platinum hypersensitivity is frequently an issue. One approach used is to desensitise people after they have a reaction. A group is looking at the possibility of prophylactically moving people to a desensitisation schedule before they have had a reaction (O’Cearbhall, Gynecologic Oncology 2010). A randomised, prospective trial on this has started.
Among 77 patients re-treated with carboplatin, 17 per cent had a hypersensitivity reaction. Among 174 people who had been prophylactically converted using the desensitisation schedule, the hypersensitivity rate was 3.4 per cent. This compared to 21 per cent of patients who remained on the standard schedule. Minimising hypersensitivity would become more important as more platinum-based doublet therapy was recommended for patients with platinum-sensitive disease, Dr Sabbatini said.
It was clear that platinum-based combination therapy was the most highly responsive regimen, Dr Sabbatini concluded.