Treatment of the first locoregional recurrence
DR. BOWLES: Unfortunately, our patient had a rapid relapse at his primary site after he completed adjuvant radiation. Could you each comment on the treatment of his local relapse?
DR. WINE: Resection is the indicated treatment for a recurrence in the oropharynx. The resection likely involved a portion of the pharyngeal wall. This caused a through-and-through defect, and the free flap was required to obtain the best chance of healing in a radiated field.
DR. BOWLES: How would your surgical recommendations be different if he had received CRT initially rather than being treated with a neck dissection?
DR. WINE: Occult regional metastasis occurs in approximately 25% of irradiated necks in recurrent HNSCC. The percentage of occult regional metastases may be around 10% if the neck is N0 prior to radiation therapy and remains N0 clinically and radiographically at the time of recurrence. Thus, elective neck dissection is definitely indicated for recurrent HNSCC if initial staging is N1 or greater. When there is a recurrence at the primary site, in necks that are initially N0, evidence exists that observation may be appropriate.
DR. REDDY: Is there a role for re-irradiation (re-RT) in this setting?
DR. RABEN: First, we should say that we agree with Dr. Wine. For patients presenting with a resectable recurrence, salvage surgery is the preferred option and offers the greatest likelihood of long-term survival. While adjuvant RT following salvage surgery offers the potential for improved locoregional control, no consistent overall survival benefit has been observed with this approach. In a recent randomized trial, 130 patients with recurrent HNSCC treated with salvage surgery were randomly assigned either to treatment with re-RT to 60 Gy combined with chemotherapy or to observation; a significant improvement in disease-free survival but not overall survival was observed with immediate adjuvant therapy. Delaying re-RT after salvage resection until subsequent locoregional failure spares unnecessary treatment-related morbidity/mortality in the small subset of patients who may be cured by surgery alone.
DR. CHEN: While controversial, re-RT following locoregional failure in HNSCC has been shown in numerous single- and multi-institutional studies to be a feasible treatment option for patients with unresectable disease. Compared to chemotherapy alone, re-RT offers the potential for more durable disease control in locoregional relapse. RTOG 9610 was the first prospective multi-institutional randomized trial evaluating re-RT for the management of recurrent HNSCC or a second primary tumor (SPT) arising in a previously irradiated field. RT was undertaken to 60 Gy in 1.5-Gy twice-daily fractions (along with weekly 5-FU and hydroxyurea), with a 2-year survival of 15.2%. RTOG 9911 similarly employed twice-daily RT, concurrent with cisplatin(Drug information on cisplatin) and paclitaxel(Drug information on paclitaxel), and demonstrated a 2-year overall survival of 25.9% in patients with recurrent HNSCC or SPT. Retrospective series have demonstrated more favorable response rates and long-term outcomes with re-RT. In the recently reported experience of the Dana-Farber Cancer Institute with IMRT-based CRT ± surgery for locally recurrent HNSCC, the 2-year locoregional control and overall survival were 48% and 67%, respectively. Although the percentage of patients with HPV-positive cancers may have contributed to these more favorable outcomes even in the recurrent setting, it is not known how many patients were HPV-positive. Such studies have shown that, for a select favorable subset of patients presenting with locoregionally recurrent HNSCC, long-term survival may be an achievable goal.
DR. RABEN: Of course, improved disease control with re-RT comes at the cost of an increased risk for treatment-related toxicity, underscoring the importance of appropriate patient selection. In RTOG 9911, for example, 28% of patients had grade 4 or worse acute toxicity, including eight treatment-related deaths. Similarly, in the re-RT experience at the Dana-Farber Cancer Institute, 91% of the 35 treated patients had at least one acute grade 3 toxicity and 46% had late grade 3 toxicity. Four late deaths occurred in patients with no evidence of disease.
DR. REDDY: How do target delineation and dose/fractionation selection in the re-RT setting compare with the upfront setting? What data support the use of hypofractionation approaches for re-RT for HNSCC recurrences?
DR. RABEN: Re-RT is usually delivered only to the recurrent tumor plus margins and does not include prophylactic nodal coverage. Conformal techniques, such as three-dimensional (3D) conformal radiation and IMRT, are essential to improve target coverage while minimizing treatment volume and associated toxicity. With conventional fractionation, doses used for curative-intent re-RT should approach those employed for initial upfront treatment. Fractionated stereotactic radiosurgery (SRS)/stereotactic body radiotherapy (SBRT) has emerged as an appealing approach that can minimize the toxicity associated with re-RT. The University of Pittsburgh recently reported their experience with SBRT for recurrent HNSCC, in which 85 patients received SBRT to a mean dose of 35 Gy (range, 15 to 44 Gy)  A complete/partial response or stable disease was observed in 88% of cases, with 1- and 2-year local control of 51.2% and 30.7%, respectively. Unger et al used SRS to a median dose of 30 Gy (range, 21 to 35 Gy) in 2 to 5 fractions for re-RT, reporting 2-year overall survival and locoregional control of 41% and 30%, respectively, with a 54% complete response rate.
DR. BOWLES: Does systemic therapy have a role in the first locoregional relapse?
DR. JIMENO: Chemotherapy can play a role with concurrent re-RT. However, chemotherapy alone is palliative, and the goal at this setting, however unlikely, remains cure. As our surgical and radiation oncology colleagues stated, the management in this case appeared standard. However, before de-escalation strategies are applied in HPV-positive patients, we need long-term data confirming their appropriateness and safety.
Treatment of the second locoregional recurrence
DR. BOWLES: Sadly, we met our patient when he had a second local recurrence, which occurred less than one year after he concluded his first therapy. His disease manifested as a PET-avid lesion in the left retropharyngeal/parapharyngeal region and was deemed unresectable (Figure 1). What can be done at this point?
DR. JIMENO: This is a setting in which chemotherapy can play a significant role. Clearly, his prognosis after a second relapse is quite grim, so therapy should focus on prolonging survival and improving quality of life. For patients with very good performance status you could consider a TPF-like regimen. However, this comes with significant toxicity. Other options include a platinum-taxane doublet, platinum–5-FU doublet, or cetuximab(Drug information on cetuximab) doublet with either a platinum or taxane. Single-agent taxanes, platinums, or cetuximab could also be used. We elected to give our patient two cycles of TP since he was chemotherapy-naive. He presented with stridor at his initial visit to our clinic, and cisplatin-docetaxel relieved this symptom.
DR. REE: The follow-up PET/CT scan showed decreased fullness of the soft tissue at the left skull base, nasopharynx, and oropharynx, with the supporting secondary sign of interval resolution of the effacement of the left fossa of Rosenmüller. Despite these areas of improvement, there remained considerable abnormal tissue and metabolic activity.
DR. BOWLES: Is it useful to employ chemotherapy to reduce the size of a relapse and then attempt a resection?
DR. WINE: There is utility in reconsidering surgery after a locoregional recurrence has been downstaged with chemotherapy or radiotherapy. It has been shown that resection of HNSCC that was unresectable at initial presentation can be beneficial if margins are negative. In the setting of locoregional recurrence, surgery may be reconsidered after downstaging, although the risks are higher after radiation (or re-RT). Because surgery would have more morbidity in this setting, the ability to achieve negative margins is of utmost importance when reconsidering resection. Thus, surgery should be considered only if there has been a significant response to chemotherapy or radiotherapy and it appears very probable that negative margin status can be achieved.
Role of Biologic Agents for Head and Neck Cancers
DR. REDDY: Our patient had not received re-RT at this point. We discussed re-RT and re-CRT earlier but we did not address a potential role for biologic agents in this setting. Are there data to support the use of targeted biologic agents, such as cetuximab, concurrently with re-RT for locally recurrent HNSCC?
DR. CHEN: Molecularly targeted agents, including cetuximab and erlotinib (Tarceva), have demonstrated feasibility and acceptable levels of acute toxicities when administered concurrently with re-RT for recurrent HNSCC. Heron et al recently reported the results of a retrospective matched-cohort study of patients with locally recurrent HNSCC, in which SBRT alone was compared to SBRT with weekly cetuximab infusion. The addition of cetuximab to SBRT improved overall survival from 14.8 months to 24.5 months, without a significant increase in grade 3/4 toxicities. A phase I dose-escalation trial of concurrent and maintenance erlotinib and re-RT at UCCC demonstrated the safety and feasibility of re-RT to 66 Gy in 30 fractions with concurrent daily erlotinib up to a dose of 150 mg daily, with a favorable acute toxicity profile compared with trials utilizing re-RT and cytotoxic chemotherapy, as well as similar late toxicity/treatment-associated mortality. Although no randomized trials to date have compared the efficacy of targeted biologics concurrent with re-RT to that of traditional chemotherapy, the above studies demonstrate that such an approach in HNSCC patients certainly merits consideration. Future trials might incorporate DNA repair inhibitors as a way to enhance radiosensitivity.
DR. RABEN: Following 2 cycles of TP chemotherapy, we elected to treat this patient with SBRT to 25 Gy delivered in 5 fractions (Figure 2), with sensitizing cetuximab during therapy and for a month afterwards. He had stable disease during treatment, but his disease progressed 2 months after he finished therapy.
DR. BOWLES: Dr. Jimeno, what novel options are available to patients with locally recurrent or metastatic HNSCC?
DR. JIMENO: There are a number of completed or ongoing trials looking at systemic therapy for locally recurrent or metastatic HNSCC. Drugs targeting EGFR have been investigated with mixed results. A phase III study showed that panitumumab plus 5-FU was no better than cisplatin plus 5-FU. Similarly, zalutumumab, another humanized anti-EGFR antibody, showed improved progression-free survival but not overall survival compared with best supportive care in platinum-refractory HNSCC. In contrast, the EGFR-HER2 inhibitor BIBW 2992 showed improved clinical benefit compared with single-agent cetuximab. We are conducting exciting phase I/II studies with PI3K inhibitors and cetuximab or conventional chemotherapy in an effort to target multiple mechanisms and thereby prevent trans-activation or bypass of key signaling pathways. We believe these and similar studies represent the next wave of rational combination strategies aimed at improving outcomes for patients with HNSCC.
Management of head and neck squamous cell carcinoma in the primary and relapsed setting requires the coordinated efforts of head and neck surgeons, radiation oncology, and medical oncology in order to maximize clinical care. Unfortunately, our patient died 4 months after completing re-RT and only 16 months after he underwent his initial surgery. Multidisciplinary efforts must be strengthened and new research performed to improve the otherwise poor prognosis for patients with locally recurrent HNSCC.
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.