Phase II Study of Low-Dose Paclitaxel and Cisplatin in Combination With Split-Course Concomitant Twice-Daily Reirradiation in Recurrent Squamous Cell Carcinoma of the Head and Neck: Results of Radiation Therapy Oncology Group Protocol 9911
Purpose: Recurrent squamous cell carcinoma of the head and neck (SCCHN) or new second primary tumor (SPT) in a previous radiation field, if not curable by surgery or radiation, is almost always fatal. Chemotherapy alone yields a median survival time (MST) of no more than 10 months and 1-year overall survival (OS) of 35% at best. Concurrent reirradiation and chemotherapy is an alternative strategy. Patients and Methods: Eligibility for Radiation Therapy Oncology Group (RTOG) protocol 9911 stipulated recurrent SCCHN or SPT in a previous radiation field. Patients received twice-daily radiation (1.5 Gy per fraction bid x 5 days every 2 weeks x4), plus cisplatin 15 mg/m2 intravenously (IV) daily x 5 and paclitaxel 20 mg/m2 IV daily x 5 every 2 weeks x4. Granulocyte colony-stimulated factor was administered days 6 through 13 of each 2-week cycle. Results: One hundred five patients were enrolled from March 2000 through June 2003; 23% had SPT. Oropharynx (40%) and oral cavity (27%) were the predominant primary sites. Median prior radiation dose was 65.4 Gy. Seventy-four percent of patients completed chemotherapy. Grade 4 or worse acute toxicity occurred in 28%, grade 4 or worse acute hematologic toxicity in 21%. Eight treatment-related deaths (8%) occurred: five in the acute setting, three late (including two carotid hemorrhages). MST was 12.1 months, with estimated 1- and 2-year OS rates of 50.2% and 25.9%. Conclusion: Despite a high incidence of grade 5 toxicity, 1- and 2-year OS rates for split-course bid radiation therapy and concurrent cisplatin/paclitaxel exceed results generally seen with chemotherapy alone. Local recurrence or second primary tumor (SPT) after curative radiation therapy (RT), alone or combined with surgery, is a major dilemma in the management of locally advanced squamous cell carcinoma of the head and neck (SCCHN). Fifty percent to 60% of patients die directly as a result of locally recurrent disease.Although surgical salvage is occasionally successful, disease extent and involvement of adjacent critical structures frequently preclude it. When surgery is not feasible, therapeutic options for local recurrence are limited. Chemotherapy alone is suboptimal. Combination platinum-based regimens generally yield response rates of only 20% to 35%, and median survival time (MST) seldom exceeds 8 to 10 months. Except in nasopharyngeal cancer and other selected clinical situations, the size and location of most recurrences does not readily permit the optimal use of interstitial radiotherapy or salvage RT alone (reirradiation). However, reirradiation combined with chemotherapy may be a viable option. In treatment-naïve locally advanced SCCHN, concurrent chemotherapy and radiation have demonstrated a clear survival advantage compared with radiation alone.On the basis of pilot trials at the Universities of Alabama and Chicago (Tuscaloosa, AL, and Chicago, IL, respectively), Spencer et al mounted a phase II trial (Radiation Therapy Oncology Group [RTOG] protocol 9610) evaluating concurrent split-course twice-daily radiation, fluorouracil (FU) infusion, and hydroxyurea in patients with unresectable recurrence or SPT within a prior RT field. Patients received RT 1.5 Gy/fraction (fx) bid for 5 days every other week x4 combined with hydroxyurea 1.5 g orally (PO) and FU 300 mg/m2 intravenously (IV), daily x5, each administered before the second fx of RT. Between August 1996 and April 1999, 86 patients were accrued. Median prior RT dose was 64.8 Gy. Twenty-five percent of patients had second primary tumors. Seventy-three percent of patients received all four cycles. Grade 3 or worse mucositis occurred in 21%, and grade 3 or worse neutropenia in 24%. There were six treatment-related fatalities: four from sepsis and two from tumor hemorrhage. MST was 8.8 months, with 1- and 2-year survival rates of 41.7% and 16.9%. The 1-year survival rate was 47.8% in those who relapsed 3 or more years after prior RT compared with 36.4% in those who relapsed within 3 years Because cisplatin and paclitaxel are established radiosensitizers with proven activity in SCCHN, investigators at Fox Chase Cancer Center (FCCC; Philadelphia, PA) mounted a phase I trial combining these agents with bid RT in the salvage setting. They employed a split-course schedule to facilitate optimal integration of chemotherapy with RT, and allowed the integration of hematopoietic growth factors (HGFs) between cycles to potentially mitigate mucositis and myelosuppression and to avoid the paradoxical toxicity of concurrent RT and HGF. FCCC 96-006 established a maximum-tolerated dose (MTD) of split-course RT of 1.5 Gy/fx bid x5 days every other week for four cycles (60 Gy), in combination with cisplatin 15 mg/m2/d x5 and paclitaxel 20 mg/m2/d x5 every other week. Without HGF, the second and subsequent cycles were delayed at least 1 week almost routinely. With HGF support, 13 of 18 cycles were delivered on schedule at full dose. The rates of grade 3 or worse neutropenia and anemia were 17% and 23%, respectively. Grade 3 mucositis occurred in only 6% of patients. Of 22 patients with measurable, recurrent SCCHN, the response rate was 54%. Median locoregional freedom from progression was 10 months, and the 1-year progression-free survival (PFS) rate was 28%. MST was 9.5 months; 1-and 2-year survival rates were 41% and 27%, respectively. On the basis of these encouraging results, RTOG initiated a phase II trial formally testing this regimen in a multicenter setting. The MST and 1- and 2-year survival rates for our approach appear promising, and generally exceed results previously observed with chemotherapy alone. ECOG trials 1393 and 1395 tested chemotherapy alone in this setting. Of nearly 400 patients accrued, 153 had local disease, of whom 124 were previously treated with radiation. The 2-year survival rate in this group was only 10.5%. In addition, the results appear superior to historical RTOG controls employing split-course RT, hydroxyurea, and infusional FU, although the usual caveats regarding selection bias and possible stage migration over time apply. Toxicity on RTOG 9911 was substantial, with a relatively high incidence of treatment-related deaths, and late grade 4 or worse toxicity rates. Patients with SPTs and those whose head and neck cancers relapsed more than 36 months after prior treatment fared no better than did patients with recurrent disease and those who relapsed within 3 years. These observations are discordant with our previous experience in RTOG 9610,wherein those who relapsed more than 3 years after prior radiation had a 12% improvement in 1-year survival rate; a similar trend favored those with SPTs as opposed to recurrent disease. Sixteen percent of patients on RTOG 9911 remain alive and progression free beyond 2 years, with long-term survival, if not cure, potentially achievable. Whether reirradiation and concurrent chemotherapy are superior to chemotherapy alone can only be addressed by a properly conducted prospective phase III trial. RTOG 0421 randomly assigned patients identical to those enrolled onto RTOG 9911 to split-course bid radiation with concurrent cisplatin and paclitaxel and G-CSF support, or to three separate cisplatin-based standard chemotherapy regimens. The primary objective of this study was OS. The eligibility criteria were identical to RTOG 9911. The study, which targeted 240 patients for enrollment, had an 80% power and 0.025 type-I error (one-sided test) to detect a 33% reduction in risk of death. Unfortunately, this trial closed because of poor accrual; the history of this ill-fated effort makes it unlikely that similar trials in the future will ever be mounted. Acknowledging the preconceived notions that frequently exist, it may be inherently difficult, if not impossible to mount a trial comparing chemotherapy alone with combination chemotherapy and RT in this setting. As concurrent chemotherapy and
radiation emerge as standard treatment in newly
diagnosed, locally advanced SCCHN, the utility of
reirradiation and further chemotherapy in the salvage setting
may be compromised, especially in patients whose disease has
relapsed relatively quickly ( In addition, the MTD of reirradiation is unclear. Although older studies have suggested limited tolerance to repeat administration of radiation, more recent preclinical experiments have indicated that soft tissue can tolerate repeat doses as high as 90% of the original dose if the second treatment is applied more than 6 weeks after the first has concluded.Subsequent tissue tolerance depends on both fraction size and field size.Langloishas shown that reducing size of the reirradiation volume can substantially decrease the probability of complications; risk of toxicity is further reduced by split-course approaches and smaller fraction sizes: 1.8 to 2.0 Gy/fx. The incidence of soft tissue necrosis in various series ranges from 0% to 40%. In our series, grade 4 and 5 soft tissue necrosis occurred in less than 5%. As others have demonstrated, total reirradiation doses of 60 Gy can be given with serious toxicity rates of less than 25%. Vokes previously employed split-course radiation with hydroxyurea and infusional FU. They observed substantial local toxicity, but a relatively high locoregional response rate, and gratifying long-term survival; MST was 12 months, with 15% of enrollees disease free in the long term. In a pilot effort, investigators at the University of Alabama at Birmingham treated 35 patients with recurrent, previously irradiated SCCHN with hydroxyurea 2 g PO bid x5, and FU 300 mg IV bolus daily x5, weeks 1, 3, 5, and 7 in combination with bid RT to a total dose of 60 Gy. Median prior RT dose was 62.8 Gy. Fourteen (40%) of 35 had previously received chemotherapy. Because of significant myelosuppression in the first 11 patients, subsequent treatment was modified, reducing RT to 1.2 Gy/fx and hydroxyurea from 2 to 1.5 g bid. As a result, hematologic tolerance improved substantially. The 1-year survival rate was 59%. However, four deaths occurred, one from aspiration pneumonia, one from neutropenic fever/sepsis, and two secondary to carotid artery rupture. The subsequent RTOG effort (RTOG 9610) employed a treatment schema virtually identical to the last cohort of the Alabama series; 60 Gy RT was administered over 4 treatment weeks (7 elapsed weeks). Acute grade 3 or worse mucositis occurred in 21% of patients, a higher percentage than observed in the 9911 trial. Six grade 5 adverse events occurred, two caused by neutropenic sepsis; OS appeared inferior to that found in RTOG 9911. Considerable reluctance to investigate reirradiation persists. Toxicity is substantial. However, appropriate patient selection restricting this approach to patients with excellent performance status should help mitigate this concern. In addition, adequate nutrition is imperative; those with compromised nutritional integrity should receive gastronomy tubes. Outside of formal studies, concurrent chemotherapy and reirradiation are not ready for routine implementation. |
1 year). How monoclonal antibodies targeting epidermal
growth factor receptor factor into this paradigm remains
uncertain. Regardless, an increasing proportion of
relapsed patients will have received radiosensitization as
part of their initial treatment.