Choice of Doublet Therapy
In the landmark ECOG 1594 study, three cisplatin(Drug information on cisplatin)-based doublets (gemcitabine, docetaxel, paclitaxel(Drug information on paclitaxel)) and carboplatin(Drug information on carboplatin)-paclitaxel were evaluated in the front-line setting for advanced NSCLC. A total of 1,155 eligible patients were enrolled, with a median age of 63 years. The median OS for all groups was 8.0 months, with a one-year survival rate of 34% and a two-year survival rate of 12%. No significant differences in OS and response rates were found between the groups. Although time to progression was improved with cisplatin-gemcitabine vscisplatin-paclitaxel (4.2 months vs 3.4 months), it occurred at the expense of more grade 4 platelet toxicity and grade 4/5 renal toxicity.
An age-based subgroup analysis was not performed in this study, but patients with ECOG PS of 2 demonstrated high rates of adverse events and were later excluded from eligibility to the study. The median OS for these patients was 3.9 months. compared to 10.1 and 7.1 months for patients with better PS (ECOG PS 0 and 1, respectively). Carboplatin-paclitaxel was chosen as the future ECOG reference regimen, due to its comparable activity (survival) and better safety (less toxicity) versus the other regimens tested (Table 1).
Belani and Fossella evaluated efficacy and toxicity of cisplatin-docetaxel, carboplatin-docetaxel, and cisplatin-vinorelbine in patients ≥65 in a subgroup analysis of the TAX 326 trial. Of the 1,218 patients enrolled, 33% were >65 years of age. Median survival was 12.6 months in the docetaxel(Drug information on docetaxel)-cisplatin arm, compared to 9.9 months in the vinorelbine-cisplatin arm and 9.0 months in the docetaxel-carboplatin arm. Elderly patients tolerated chemotherapy relatively well but experienced increased neurotoxicity, asthenia, pulmonary toxicity, diarrhea (8.1% vs 5.8%), neutropenia (81.8 vs 70.7%), and febrile neutropenia (8.1% vs 3.1%).
Clinically speaking, patients in the TAX 326 trial and other trials may have better tolerance if therapy is dosed weekly, as described elsewhere,[22, 23] this is possible because weekly dosing may allow for more frequent surveillance of toxic adverse effects, dehydration, and infections during therapy (Table 1).
Ansari et al presented a study in which a retrospective subgroup analysis of the elderly (>70 years old) was performed on a randomized phase III study comparing carboplatin-gemcitabine, carboplatin-paclitaxel, and gemcitabine(Drug information on gemcitabine)-paclitaxel doublets in metastatic disease. Data were pooled from all three treatment arms to evaluate efficacy and toxicity in older versus younger patients. No significant differences were found in response rates, PFS, or OS. Slightly more hematologic toxicity was reported in the ≥70 population (Table 1).
The introduction of pemetrexed(Drug information on pemetrexed)-based combination chemotherapy has altered standard practice for first-line therapy. In 2008, Scagliotti et al compared cisplatin-pemetrexed vs cisplatin-gemcitabine in a randomized phase III study of patients with PS 0-1 receiving first-line treatment for metastatic or unresectable NSCLC. The median age in the study was 61 years; 37.2% of cisplatin-pemetrexed and 33.1% of cisplatin-gemcitabine patients were ≥65 years. In a pre-planned analysis, the use of cisplatin-pemetrexed in non-squamous histology was associated with a statistically significant improvement in both OS and PFS compared to cisplatin-gemcitabine (11.8 vs 10.4 months and 5.3 vs 4.7 months, respectively). This study noted the importance of histology when selecting pemetrexed-based therapy and compared favorably with historical controls with respect to OS.
In patients ≥65 years old, the survival hazard ratio trended in favor of cisplatin-pemetrexed (0.88). Significantly more grade 3/4 hematologic, febrile neutropenia, alopecia, and red blood cell and platelet transfusion toxicities were seen in the cisplatin-gemcitabine group. More grade 3/4 nausea was seen in the cisplatin-pemetrexed group. In summary, cisplatin-pemetrexed demonstrated improved OS and PFS in non-squamous histology patients, with less toxicity. There was a trend toward increased survival in the elderly (Table 1). Updated survival data has resulted from pemetrexed maintenance therapy given after induction platinum doublet chemotherapy, and this has taken advanced NSCLC survival to new levels. The OS for the entire cohort of patients in this study was 13.4 months, and the number was even higher in the subgroup of patients with adenocarcinoma histology.
Overall, the use of any platinum doublet to treat advanced NSCLC shows similar efficacy, but with slight differences in toxicity. Prior to the introduction of pemetrexed, which is more effective in non-squamous histologies, no specific clinical characteristic would aid in the selection. Together, the above-mentioned studies continue to reveal an ongoing theme, demonstrated mostly with post hoc analysis, retrospective subgroup analyses, and rarely in randomized trials: Elderly patients have similar levels of response rates and OS, but added toxicity, particularly myelosuppression.
Physicians who treat adenocarcinoma and non-squamous histology have added options to their treatment regimens that include pemetrexed therapy. Whether the survival benefit from pemetrexed is best achieved via usage as a front line, second line, or maintenance is yet to be determined. This remains a viable and relatively well-tolerated agent for elderly patients who require treatment for advanced-staged disease. The choice of therapy for advanced NSCLC with squamous histology still remains difficult as our therapeutic armamentarium is limited, regardless of age. Reasonable options include platinum/taxane combinations or combinations with gemcitabine in this group of squamous histology patients.
In patients with unresectable stage III NSCLC, the treatment of choice has often been combined modality chemoradiotherapy. As in the past, the most current ASCO guidelines suggest that a concurrent chemoradiotherapy approach be reserved for patients with good PS. As clinicians, we are all well aware of the toxicities, particularly esophagitis, pneumonitis, and myelosuppression, that are related to concurrent chemotherapy and how they contribute to morbidity and mortality of patients, even in our youngest and fittest patients.
There are no prospective randomized trials that have evaluated chemoradiotherapy in elderly patients. Prospective trials that include elderly patients and that were designed to evaluate the role of combined therapies commonly employed in clinical practice are detailed in Table 2. Modifications in practice of the regimens listed in Table 2 often include weekly dosing, substitution of cisplatin for carboplatin, and reduction in total days of etoposide(Drug information on etoposide) administered.
Studies completed by Albain et al and Ganadra et al, which utilized cisplatin-etoposide concurrent with thoracic radiation, included only patients with ECOG PS of 0 or 1 and limited numbers of elderly patients. Although these Southwest Oncology Group regimens have shown impressive median OS data of 26 months, the resulting short-term hematological and non-hematological toxicities may outweigh the benefits in less-fit patients.[28,29]
There are several retrospective subgroup studies taken from randomized phase III trials that evaluated concurrent chemoradiotherapy between patients ≥70 years old and those <70 years old, with only one trial (Sgroi et al) evaluating consolidation chemotherapy.[30-32]. Survival results were similar in elderly and younger patients; however, the former experienced increased esophagitis, myelosuppression, and dehydration and a higher rate of early treatment termination. Altogether, combining the previous prospective phase II data (Table 1) and the hypothesis-generating data from these retrospective reviews, we can clearly establish that elderly patients must have their therapies tailored to PS with careful observation and quick intervention for the increased toxicity they will experience during chemoradiotherapy.
Targeted Therapies for Metastatic Disease
In the ECOG 4599 randomized phase III trial, the group assigned to paclitaxel-carboplatin (PC) plus bevacizumab(Drug information on bevacizumab) (PCB) demonstrated a median survival of 12.3 months compared with 10.3 months in the chemotherapy-alone (PC) group (hazard ratio for death: 0.79; P = .003). A subgroup analysis of this study evaluated 224 patients aged ≥70 (26%) from the total 878 initial trial participants. In terms of response and PFS, elderly patients treated with PCB (compared to PC alone) showed a trend toward higher response rates (29% vs 17%; P = .067) and PFS (5.9 vs 4.9 months; P = .063). There was no difference in OS (11.3 vs 12.1 months; P = .40); however, this subgroup analysis was not designed to detect such a difference. Elderly patients had higher incidence of grade 3-5 adverse events in the PCB than in the PC alone arm (87% vs 61%; P < .001). Neutropenic fever rate was 6.2% (PCB) vs 0.9% (PC) (P = .03). Elderly patients on the PCB arm also had more bleeding, anorexia, and nausea than did younger patients. Most importantly, among elderly patients there were seven treatment-related deaths on PCB, versus only two deaths on the PC alone arm. Further study is warranted in order to understand which elderly patients will benefit from, and which will be harmed by bevacizumab.
The Canadian BR.21 study evaluated erlotinib versus placebo in the second line and reviewed outcomes of 163 (22%) elderly patients (age ≥70 years) out of 731 total patients. There were no significant differences between elderly and non-elderly patients regarding PFS (3.0 vs 2.1 months for erlotinib vs placebo; P = .009), response rates, and OS (7.6 vs 5.0 months in elderly [P = .67] and 6.4 vs 4.7 months in young patients [P = .0014] for erlotinib vs placebo, respectively). Elderly patients also derived similar quality-of-life improvements from treatment. Regarding toxicity, more elderly patients had grade 3/4 toxicity (35% vs 18%; P < .001), and more elderly patients were required to stop treatment because of these toxicities (12% vs 3%; P < .001). Rash and diarrhea were the most common adverse events. Thus, erlotinib remains a viable treatment option among elderly patients for second-line therapy.
NSCLC is becoming increasingly common among our elderly population. The most important factor in choosing a patient’s initial and subsequent chemotherapy is PS, not age. We obviously must be cautious in our choice of therapies, as decreasing bone marrow and kidney reserve, hearing loss, reduced hepatic synthetic function, and corresponding tobacco-related pulmonary and cardiovascular complications can lead to increased therapy-related toxicities in the elderly population. In one of the few randomized trials performed in elderly patients with advanced NSCLC, Quoix et al clearly demonstrated a survival benefit for doublet chemotherapy with carboplatin and paclitaxel over single-agent gemcitabine or vinorelbine. Therefore, offering single-agent therapy because of age alone is no longer justified. Prior to the introduction of pemetrexed for non-squamous histologies, no survival benefit existed among platinum doublets for advanced disease; however, elderly patients consistently suffered increased toxicity. The use of pemetrexed, when appropriate, offers improved survival and less toxicity, a benefit that is retained among the elderly. There is no standard of care for concurrent chemoradiotherapy in this population.
A concerning and noteworthy observation that has come from retrospective subgroup analyses is that grade 5 events associated with bevacizumab have been observed in the elderly. The toxicity profile of different biologic agents in the elderly is a field of great interest that requires further study both clinically and in the laboratory.
The efficacy of individualized cancer care—based on a patient’s tumor genetics and mutational status—with agents such as erlotinib and other targeted therapies has yet to be evaluated in this elderly population. Cost effectiveness also becomes a question as we look toward an era in which EGFR, KRAS, ALK, and others, are evaluated during diagnosis. Oncologists and insurers will have to balance their excitement over novel discoveries with a rational assessment of the level of tolerance they see when implementing new treatments in our elderly population. Advanced NSCLC is the same disease in young or old, and treatment decisions must be based on PS and comorbidities generating specific pharmacologic contraindications. All oncologists must discuss the risks of increased toxicity with their elderly patients and caregivers and inform them of how to best deal with the potential complications, should they arise.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
Acknowledgments: We thank Rasa Hamilton (Moffitt Cancer Center) for editorial assistance.