Experts provide an in-depth look at the treatment strategy in patients with solid tumors, specifically looking at the role of maintenance therapy.
Dennis Citrin, MD, PhD
Ten years ago, Dennis Citrin, MD, PhD watched as patients with metastatic non—small cell lung cancer died, despite treatment, 6 to 12 months after diagnosis.
These days, through the use of chemotherapies (eg, carboplatin, paclitaxel, pemetrexed) and targeted treatments (eg, bevacizumab) the doctor is able to usher many of them into remission.
It’s a transformation that has, in turn, created the need for longer-term treatment strategies—not just for patients with NSCLC, but for those with other solid tumor types in which prognoses have improved, said Citrin, MD, Phd, of the Cancer Treatment Centers of America.
Increasingly, the oncology community is looking to meet that need with maintenance therapy (MT), a strategy that involves immediately following a successful first-line treatment with a component of the same regimen (continuation maintenance), or with a different therapy that is effective and well-tolerated (switch maintenance).
“When it comes to more common solid tumors like breast, colorectal, or ovarian, the goal of initial treatment is to get the disease under control so it’s not causing symptoms or discomfort and, more than that, to prolong survival,” said Citrin, who practices at his organization’s Midwestern Regional Medical Center in Zion, Illinois. “Having said that, you can’t keep patients with these cancers in remission without continued treatment, and that’s the whole idea behind maintenance therapy.”
Heather Wakelee, MD
Without some of the newer, less-toxic cancer medications, MT wouldn’t be a workable strategy, particularly in NSCLC, noted Heather Wakelee, MD, assistant professor of Medicine in the Division of Oncology at Stanford University, in California.
“Why haven’t we done maintenance before? It’s because we were focused on platinum-based doublet chemotherapy in treating NSCLC, and you can’t give doublets indefinitely, nor can you give taxanes indefinitely,” Wakelee said. “It’s only because we have drugs that you can continue, from a toxicity standpoint, that this question has really come up.”
MT, which can include cytotoxic drugs, targeted therapies, hormonal or other anticancer treatments, has long been established in blood cancers, including acute lymphocytic leukemia and acute myeloid leukemia; once in remission after induction therapy, those patients get MT for 2 or 3 years to protect against their cancer’s return.1
Debu Tripathy, MD
While the concept of MT remains under investigation in many types of solid tumors, the strategy has been adopted for eligible patients in some breast cancers—although doctors don’t always refer to it as MT, said Debu Tripathy, MD, co-leader of the Women’s Cancers Program and a professor of Medicine at the Norris Comprehensive Cancer Center at the University of Southern California in Los Angeles.
In early-stage breast cancers, where cure is the goal, patients often undergo a fixed number of chemotherapy cycles, followed by what some consider maintenance with hormonal drugs or trastuzumab for a prespecified length of time. In advanced or metastatic disease, where the aim is to slow disease progression, there are fewer rules for what a regimen should contain or how long it should last, except that both initial and maintenance therapies usually continue until disease progression, Tripathy added. In that setting, maintenance treatments vary based on a tumor’s size, its responsiveness to treatment, and the severity of the side effects a patient experiences, Tripathy said.
“We use maintenance therapy as part of standard practice, but we haven’t done many trials to say when to optimally introduce it or when to use new drugs, and it’s still controversial at what point we discontinue any of the therapies,” he said. “Furthermore, we don’t really have drugs that we use purely in the maintenance therapy setting, other than the biological drugs. Determining when to use maintenance therapy still remains part of the art of oncology. You have to use your common sense as best you can, and use the data we have to decide when and how and in what situation to use maintenance therapy.”
The particulars of MT are more defined in nonsquamous NSCLC. In patients with advanced NSCLC who have stable disease, good performance status, and other favorable factors, MT should be given after 4 to 6 cycles of first-line chemotherapy, according to the National Comprehensive Cancer Network’s (NCCN’s) 2013 Clinical Practice Guidelines in Oncology. Drugs approved by the FDA for maintenance treatment of such patients are pemetrexed and erlotinib, and the NCCN also recommends bevacizumab and gemcitabine in appropriate patients. Docetaxel is also sometimes used in NSCLC MT, according to Marie-Anne Smit, MD, MS, and John L. Marshall, MD, in their recent review of MT in various types of solid tumors.2
Despite established standards, debate about the value and role of MT in NSCLC continues. Some argue that more evidence of the efficacy of MT is needed, adding that the tactic is costly and can increase toxicities for patients.
Corey Langer, MD
But Corey Langer, MD, director of Thoracic Oncology at the Abramson Cancer Center and professor of Medicine at the University of Pennsylvania, is among those who tend to favor the strategy because its immediacy prevents treatment attrition following first-line therapy.
In Langer’s opinion, the toxicity associated with maintenance pemetrexed is minimal, and he’s seen it work well in patients.
“My record is a woman who’s on her 50th cycle with pemetrexed maintenance given every three weeks. That’s three years altogether— uncharted territory,” he said. “I hardly ever remember advanced lung cancer patients making it to a year or two, so having a debate at three or four years about whether to continue treatment is a nice dilemma.”To Mark Kris, MD, chief of the Thoracic Oncology Service at Memorial Sloan-Kettering Cancer Center, there’s no question that MT has value in NSCLC.
“To me, it’s almost a no brainer that continuing a therapy that has proven to be effective and safe in a specific patient makes all the sense in the world—it’s the ultimate targeted therapy,” Kris said. “It gives such patients a longer time cancer-free and an improvement in survival measured in months. For those people in whom it’s safely given, effective, and not lifestyle-impeding, all the data from virtually every trial in NSCLC says it should be continued.”
Mark G. Kris, MD
“The PARAMOUNT trial3 that looked at maintenance therapy with pemetrexed—regardless of whether you measure the time from the start of treatment or the beginning of maintenance— demonstrated months of improved survival,” Kris said. “In POINTBREAK,4 although overall survival (OS) was the same, when you look specifically at patients who got maintenance therapy, giving them both pemetrexed and bevacizumab, rather than bevacizumab alone, resulted in longer survival, as well.”
In a debate during the 13th International Lung Cancer Congress in July 2012, Wakelee echoed Kris’s argument that MT is a valuable tool for managing NSCLC.
Wakelee described switch maintenance as a standard in the disease, noting that a progression-free survival (PFS) benefit has been demonstrated in all studies of that approach in NSCLC. In addition, she said, OS improved significantly with pemetrexed switch maintenance in the JMEN trial5 and with erlotinib switch maintenance in the SATURN trial,6 and strongly trended in favor of docetaxel as MT rather than second-line therapy in Fidias et al’s phase III study.7
pemetrexed maintenance vs placebo, following four cycles of pemetrexed plus cisplatin
Median OS, 13.9 vs 11.0 months from randomization, 16.9 vs 14.0 months from induction, with pemetrexed vs placebo (HR = 0.78; P = .02)
Median PFS, 4.1 vs 2.8 months from randomization, with pemetrexed vs placebo (HR = 0.62; P <.0001)
pemetrexed/carboplatin/bevacizumab, followed by pemetrexed/bevacizumab maintenance
paclitaxel/carboplatin/bevacizumab, followed by bevacizumab maintenance
Overall: Median OS, 12.6 vs 13.4 months in pemetrexed vs paclitaxel arm (HR = 1.00; P = .949)
Median PFS, 6.0 vs 5.6 months in pemetrexed vs paclitaxel arm (HR = 0.83; P = .012)
Subgroup treated with maintenance:
Median OS, 17.7 vs 15.7 months in pemetrexed/bevacizumab vs bevacizumab arm
pemetrexed maintenance vs placebo, following four cycles of platinum-based chemotherapy
OS, 13.4 vs 10.6 months, with pemetrexed vs placebo (HR = 0.79; P = .012)
PFS, 4.3 vs 2.6 months with pemetrexed vs placebo (HR = 0.50; P <.0001)
erlotinib maintenance vs placebo, following four cycles of platinum-based chemotherapy
Overall: Median OS, 12.0 vs 11.0 months with erlotinib vs placebo (HR = 0.81; P = .0088)
Median PFS, 12.3 vs 11.1 weeks with erlotinib vs placebo (HR = 0.71; P <.0001)
Median PFS with erlotinib vs placebo: HR = .10; P <.0001
Fidias et al7
maintenance vs second-line docetaxel, following four cycles of gemcitabine/carboplatin
Median OS, 12.3 vs 9.7 months, with maintenance vs second-line docetaxel (P = .0853)
Median PFS, 5.7 vs 2.7 months, with maintenance vs second-line docetaxel (P = .0001)
Continuation maintenance, meanwhile, is standard in the disease with bevacizumab, cetuximab, and erlotinib, and there is growing evidence of value with gemcitabine, and possibly with pemetrexed, Wakelee said.
When oncologists don’t intend to continue a specific drug regimen after induction therapy, Wakelee noted, they often face a choice of giving a patient either switch maintenance or a treatment holiday, followed by second-line therapy. While many patients would fare as well with either approach, she said, there is a strong rationale for choosing continuous treatment.
“Forty percent of patients who go onto first-line trials and are then eligible for second-line therapy never get it,” Wakelee said. “They do OK, have a chemotherapy holiday, and never get to the next line of therapy. [Ultimately,] these patients don’t do as well. Studies show that this is true around the world, even at the top centers.”
In Fidias et al’s trial that favored docetaxel as MT rather than second-line treatment, 95% of patients eligible for MT received it, while only 63% of patients eligible for second-line therapy received treatment, Wakelee pointed out.
Despite those data, Langer said, when “patients have a cushion of time before they need additional treatment, a lot just want to take a break. They have a low disease burden and they want time to reconstitute themselves, go on vacation, and recover from platinum-based toxicities.” And oncologists, he said, are often comfortable going along with that.
But by choosing that route, patients risk losing their chance to get second-line treatment due to “variations in performance status, the frequency and nature of clinical and radiographic follow-up, local practice patterns, and the possibility of rapid clinical decline precluding further treatment,” wrote David E. Gerber and Joan H. Schiller in their 2013 paper on MT in NSCLC.8
The authors added that continuation maintenance chemotherapy is preferable to second-line chemotherapy because the early use of non—cross-resistant therapies may kill more cancer cells prior to the development of resistance, and because continuous, low-dose chemotherapy has antiangiogenic and immunomodulatory effects.
“We never stop second- or third-line therapy while it’s working,” Wakelee argued, “so why do we ever think about stopping first-line?”
She added that there isn’t much to gain by giving patients a treatment break.
“The argument that patients need a chemotherapy holiday to feel better is not a strong one,” she said. “Patients tend to feel better when they’re on a chemotherapy that’s working.”Offering a counterpoint during the debate at the Lung Cancer Congress was Glenwood Goss, MD, FCPSA, FRCPC, professor of Medicine and director of Clinical and Translational Research at the Ottawa Hospital Cancer Centre at the University of Ottawa.
Glenwood Goss, MD
Goss noted that, in trials of MT in NSCLC, there are many patients who never make it to the maintenance phase.
In studies of MT for NSCLC, he said, just 30% to 60% of patients who underwent induction therapy were then deemed eligible to receive maintenance. Of those patients, the ones randomized to placebo were more likely, after the study was over, to be eligible for treatment upon progression, Goss said, “meaning that maintenance therapy comes at a cost.”
Goss added that he hasn’t seen convincing evidence that MT works in the disease. In continuation maintenance, he said, PARAMOUNT was the only trial among six large, recent studies to demonstrate a statistically significant survival advantage associated with that approach. Similarly, he said, among three trials of chemotherapy switch maintenance in NSCLC, the JMEN study was the only one to show a statistically significant survival advantage, and SATURN stood alone among six phase III trials of targeted MTs in NSCLC in demonstrating a statistically significant improvement in OS.
In the PARAMOUNT study, which Goss described as “our most recent, and one of our best,” just half of the 1000 patients screened went on to get MT, he said. Then, he said, the treatment didn’t do as much as he would have liked to bolster OS.
“The median survival in the pemetrexed arm was 16.9 months, and the median survival in the placebo arm was 14 months, so what the patient undergoes is 3 months of treatment for 2.9 months of improvement in survival—they get treated longer than the increase in survival,” he said.
Moreover, of four meta-analyses that assessed the success of trials of MT in NSCLC,9,10,11,12 only one showed a statistically significant increase in PFS regardless of maintenance type, and in that case the hazard ratio was 0.92, Goss said—“rather minimal, if you ask me.”
Two of the meta-analyses reported a statistically significant improvement in OS with switch maintenance but not with continuous, and one demonstrated no difference in OS, Goss reported.
Goss argued that “no trial has demonstrated an improvement in quality of life with MT versus observation—not one,” although some showed no deterioration in quality of life. In addition, he said, three of the meta-analyses showed an increased adverse event (AE) rate in the maintenance arm; the fourth didn’t evaluate AEs.
Finally, Goss said, in the majority of cases, it’s not clear which patients are likely to benefit from MT.
“We desperately need good biomarkers of efficacy if we’re going to persist with this strategy,” he said.
According to Gerber and Schiller, though, there are already some predictors that can help identify which patients will benefit from MT.
Depending on the type of drug patients with NSCLC are scheduled to receive, the authors wrote, those factors can include stable disease after chemotherapy; good performance status; potential obstacles to receiving second-line therapy at disease progression; adenocarcinoma, rather than squamous cell, histology; and EGFR mutations.
Langer added that there might be stronger evidence to support the value of MT in lung cancer if trials had been designed differently.
“I think the reason there’s so much criticism, and that you haven’t seen wholesale adoption of maintenance therapy, is the lack of mandatory crossover of the control arm to the maintenance therapy drug at progression,” which would have protected against inadvertent bias, Langer said. “That was done in the Fidias trial [immediate versus delayed docetaxel],” Langer said, “and if it had been done in the Ciuleanu paper [maintenance pemetrexed versus placebo] and we had still seen a survival advantage, the use of maintenance therapy would have gone up.”The use of MT in breast cancer is based on studies, over the last 10 years, that have shown it delays disease progression, Tripathy said. However, he cautioned, the studies have found little evidence that MT increases OS for patients with breast cancer.
According to Smit and Marshall in their review of MT in solid tumors, the strategy is standard in two types of breast cancer.
Anti-estrogen or aromatase inhibitor continuation therapy is the go-to strategy for patients with estrogen receptor (ER)— and/or progresterone receptor (PR)–positive metastatic breast cancer (MBC), they wrote. The technique, they specified, is often used after the cancer has been treated with surgery and chemotherapy and no evidence of disease remains.
According to Tripathy, such adjuvant therapy is not typically considered to be MT. But a similar tactic, in cases where ER- or PR-positive breast cancer is growing rapidly or has affected organs, does fall into the MT category, he said. Patients with that kind of disease, he said, might start with chemotherapy and then be given switch maintenance with up to 5 years of a hormonal therapy, such as tamoxifen, for premenopausal women or an aromatase inhibitor for postmenopausal women.
Meanwhile, in advanced HER2-positive breast cancer, the typical treatment is 3 or 4 months of chemotherapy with trastuzumab, and then, if the tumor is shrinking and the patient is experiencing side effects, maintenance with trastuzumab alone until disease progression or side effects that interfere with quality of life, Tripathy said.
Finally, if a tumor in advanced breast cancer is both ER-negative and HER2-negative, the only available treatment is chemotherapy, which can continue as long as the benefits outweigh the side effects, Tripathy said. “Sometimes, if a patient is on two chemotherapies, we might drop one and continue the other, and some people call that maintenance therapy,” he said. “It’s a matter of semantics.”
In breast cancer, MT with chemotherapy is much less common than maintenance with other types of agents, added Tripathy, who described the practice as “controversial.”
“There have been some studies with MT chemotherapy in breast cancer,” he said, “but it’s not like in lung cancer, where we have a series of trials that have shown that it clearly improves outcomes and the FDA has approved chemo in the MT setting.”
Maintenance with chemotherapy is being studied in patients with HR-negative or rapidly progressing MBC, or with visceral or endocrine therapy— resistant disease, Smit and Marshall pointed out. Several agents have been evaluated, the colleagues wrote, including pegylated liposomal doxorubicin (PLD), IL-2 with 13-cis retinoic acid, paclitaxel, and capecitabine, with only PLD demonstrating an improvement in time to progression (TTP) in the phase III GEICAM 2001-01 trial.13
That trial compared PLD with observation in patients who had initially been treated with doxorubicin and docetaxel. In the PLD arm, TTP was significantly improved by 3.3 months (8.4 vs 5.1 months; hazard ratio = 0.54; P = .0002), but a modest increase in survival was not significant.
At the 2012 ASCO Annual Meeting, continuation maintenance with paclitaxel and gemcitabine was also reported to benefit MBC patients. In a phase III trial known as KCSG-BR 0702,14 investigators looked at MT with the combination versus observation in MBC patients who had responded to first-line treatment with the two drugs. From the time of randomization to disease progression, patients experienced superior median PFS in the experimental versus the control arm, with a 27% improvement from 3.8 to 7.5 months (P = .026), and OS rates that were 35% better in the maintenance arm, with a nearly 9-month improvement in survival for patients who continued treatment (P = .048). The authors of the study suggested the course of treatment for patients with HR-negative, visceral MBC with a high tumor burden.
Alessandra Gennari, of Galliera Hospital in Genoa, Italy, and coauthors provided an overview of the value of maintenance chemotherapy in MBC in 2011.15 In a meta-analysis, Gennari et al evaluated the results of 11 randomized clinical trials that included a total of 2269 patients, with a focus on the duration of first-line chemotherapy regimens and patient outcomes. Some of the regimens involved continuation or switch maintenance. “Longer first-line chemotherapy duration resulted into a marginally longer OS (hazard ratio = 0.91; P = .046) and a substantially longer PFS (hazard ratio = 0.64; P <.001),” Gennari et al wrote.MT is also being employed in other solid tumor types.
In superficial bladder cancer, maintenance intravesical therapy is standard treatment, according to Smit and Marshall. And in both ovarian and colorectal cancers, they wrote, MT has demonstrated promising results in phase III trials.
Specifically, they said, two phase III trials16,17 tested continuation MT with paclitaxel in ovarian cancer, doubling PFS from 12 to 24 months (P = .016) and improving OS from 38 to 80 months (P = .012). In colorectal cancer, they wrote, a chemotherapy discontinuation trial used continuation maintenance with leucovorin and fluorouracil, improving PFS from 6.6 to 8.6 months (hazard ratio = 0.61; P = .0017).
Phase II trials have generated promising results in even more cancer types, including malignant pleural mesothelioma, pancreatic and biliary tree carcinoma, carcinoma of any primary site, head and neck cancer, melanoma, high-grade glioma, and urothelial cancer, Smit and Marshall wrote.
Based on all those results, there’s a growing need for randomized clinical trials of MT, according to Smit and Marshall, with endpoints not just of OS and PFS, but also of quality of life and toxicity.
As such research moves forward, Smit and Marshall added, investigators should help make their studies comparable by adopting specific, universal definitions of MT.
“Although MT is being increasingly investigated in the setting of metastatic solid tumors, it is becoming an established concept only in a few,” they wrote. “With agreement in terminology and with trials designed to specifically evaluate the effect of MT, future trials will be able to better address the role of maintenance therapy in metastatic solid tumors.”With MT playing a growing role in the mix of cancer treatments, both benefits and cost implications can be expected, said Jan. E. Berger, MD, a healthcare consultant and founder of Health Intelligence Partners.
Jan E. Berger, MD
“Value is defined as an equation of quality over cost,” Berger said. “If patients are active, working, bringing money into the national coffers, and paying taxes, the value of long-term treatment is pretty clear. If it’s a matter of going on a chronic medication to give you 4, 6, or 8 months longer of life, that’s the dilemma we’re trying to figure out today in cancer care.”
American health insurance companies, which decide what medical services they are willing to pay for, are likely to support therapies that are recommended in the clinical guidelines of medical advisory associations, such as the NCCN, Berger said. However, she said, the insurers might then compensate for those costs by reconsidering their support of other covered services or products.
Even for patients, decisions about the value of MT can come down to money: how much to spend on expensive medications in the near term versus how much to save for family, Berger said.
In his debate about MT in NSCLC, Goss said that maintenance can come with a “significant” price tag. The incremental cost-effectiveness ratio (ICER) per life-year gained for maintenance pemetrexed in patients with nonsquamous histology, as compared with observation, is $122,371, he said, referring to a study by Klein et al,18 the first to evaluate the cost-effectiveness of MT in advanced NSCLC.
In some cases, comparable medications can come at vastly different costs, Gerber and Schiller added. They pointed to a study that compared the costs of maintenance paclitaxel and maintenance bevacizumab in ovarian cancer—both of which have been shown to substantially improve PFS—following initial therapy with carboplatin and paclitaxel. The first-line regimen plus maintenance paclitaxel had an ICER of $13,402 per quality-adjusted life-year, while the first-line regimen followed by maintenance bevacizumab raised the ICER to $326,530 per quality-adjusted life-year, they wrote.
Berger agrees that the costs of maintenance pemetrexed and bevacizumab might be considered high by payers. In countries whose governments use equations such as ICERs to decide which health services to fund, she said, such price tags could be deal breakers.
“If the quality cost per life-year is over $50,000 or $100,000,” Berger said, “they don’t cover it.”
Limits tend to be similar among American payers, but should be increased, according to Klein et al.
“The base case cost per life year gained (LYG) for maintenance therapy with pemetrexed plus best supportive care (BSC) compared with observation plus BSC may not be considered cost-effective when compared with a commonly mentioned threshold of $100,000 per LYG in the United States,” they wrote. “However, this threshold is often criticized for being outdated and not taking into consideration the severity of the condition. In addition, this threshold is not well documented in the medical literature. Braithwaite et al19 have recently developed plausible lower and upper bounds for cost-effectiveness decision rules in the United States ranging from $95,000 per LYG to $264,000 per LYG. Based on these revised thresholds, in a population with advanced, nonsquamous NSCLC, maintenance therapy with pemetrexed is cost-effective compared with observation or erlotinib.”
To better determine which therapies deserve coverage, Berger suggested, clinical trial investigators should use patient registries—databases that follow the progress of patients over time.
“They answer questions about value, like what the side effects are, whether it’s hard for somebody to continue to take the medication, and whether patients experience depression,” she said.
In the meantime, Berger said, the costs and benefits of MT and other cancer treatments need to be part of “an active discussion,” even though they’re emotional topics.
“What a life is worth financially is a very philosophical and difficult societal conversation,” she said. “It’s a discussion we’re all afraid to have, but one we can no longer shy away from.”