Clinical Outcomes of Nasopharyngeal Carcinoma Patients... : Nigerian Journal of Clinical Practice (2024)

INTRODUCTION

Nasopharyngeal carcinoma (NPC) is a unique head and neck malignancy arising from the epithelium of nasopharynx and accounts up to 0.7/100.000 worldwide in 2018 with an approximately 2- to 3-fold male predominance.[¹] Significant variation exists between different geographical regions, Turkey, though being in the middle of two continents Asia and Europa have an incidence of 0.4 to 1.2 per 100 000 for females and males increasing after age 40.[²] Concurrent chemoradiotherapy (CRT) is the main curative approach for loco-regional disease, excluding early node-negative state (T1N0).[³] The nasopharyngeal region is adjacent to highly delicate organs such as the brainstem, optic nerves, salivary glands, and swallowing muscles, which necessitates highly conformal radiation planning. Intensity-modulated radiotherapy (IMRT) is the standard treatment of choice for NPC in terms of dose homogeneity and sparing of critical structures, which can lead to improved local control and reduced side effects.[⁴⁵]

Tomotherapy is a unique way of applying IMRT with the ability of megavoltage computed tomography (MVCT) image guidance to eliminate inter-fractional variations. Simultaneous integrated boost (SIB) technique allows different dosing for tumor and risky subclinical areas with one inverse planning.[⁶⁷] This method of IMRT has been clinically confirmed as safe and effective and eventually became the predominant mode of radiation delivered with improved dose distributions.[⁸] Increased fraction dose (2,12-2.4 Gy) to the tumor site and shortening overall treatment time can provide better local control by reducing accelerated repopulation of tumor cells.[⁹]

Adaptive RT is defined as the modification of the radiation treatment plan delivered to a patient during therapy to account for temporal changes in anatomy for example, tumor shrinkage or weight loss. It reoptimizes the treatment to avoid unnecessary doses received by critical structures such as parotid glands and reduces the risk of xerostomia.[¹⁰]

This study aims to present 5-year clinical outcomes of 67 non-metastatic NPC patients treated with adaptive SIB-IMRT technique in our helical tomotherapy (HT) unit.

MATERIALS AND METHODS

Patients: Between February 2010 and September 2017, 67 pathologically confirmed, non-metastatic NPC patients treated in a tertiary radiation oncology clinic HT unit, Ankara, Turkey were retrospectively reviewed upon approval of an ethical board. Informed consent was taken before treatment. Patient files and hospital electronic information systems were used for data collection. Direct phone calls were done when needed. All patients had detailed Head & Neck examinations, including the endoscopic evaluation and their biopsy taken either from neck nodes or nasopharynx. All patients had pretreatment magnetic resonance imaging (MRI) of nasopharynx and neck and staged mostly using Positron emission tomography-computed tomography (PET-CT) (n = 62, 92.5%) according to the tumor node metastasis (TNM)^eighth staging system.

RT planning: Patients were immobilized with a thermoplastic head and shoulder mask, and a ≤3-mm slice thickness of plain computerized tomography (CT) images were taken before treatment. IV contrast was used according to treating physician desire. Primary and nodal gross tumor volumes (GTVs) were delineated with the guidance of MRI and PET-CT fusion of simulation CT images. GTVs were expanded 5 mm isotropically to form clinical target volume 70 (CTV70), and CTV70 was expanded 3 mm to form a planning target volume of 70 Gy (PTV70). The margin between GTV to CTV can be reduced to 1 mm in adjacent critical normal structures such as the brainstem. High-risk clinical target volume of 60 Gy (CTV60) encompasses a 5- mm isotropic expansion of PTV70 and the areas at risk for microscopic involvement including whole nasopharynx, inferior sphenoid sinus, posterior parts of paranasal sinuses, base of skull, clivus, pterygoid fossa and high risk nodal levels bilaterally (retropharyngeal nodes, neck node levels 2, 3,4 absolutely, 1b if level 2 is involved, 5 if level 2,3,4 are involved). Low-risk clinical target volume of 56 (CTV56) includes the upper part of sphenoid sinus if not involved and clinical negative neck node levels 4 and 5 bilaterally. CTVs were rearranged with the exclusion of bone, air and then expanded with a 3-mm isotropic margin to form corresponding PTV60 and PTV56. PTVs were cropped 2-mm from the skin unless suspicion of dermal lymphatic layer invasion. Bilateral orbits, optic nerves, chiasma, brainstem, cochlea, temporomandibular joints, spinal cord, parotid glands, oral cavity, and esophagus were contoured as organs at risk (OAR).

For PTV70, PTV60, and PTV56, the planning dose at D95 was prescribed to 70 Gy (2.12 Gy/fx), 59.4 Gy (1.8 Gy/fx), and 56 Gy (1.7 Gy/fx) respectively in 33 fractions. Subtraction PTV volumes generated for optimization purposes. Planning parameters were: Field width (FW) = 2.5 cm, pitch = 0.215 and modulation factor (MF) = 2-3. No more than 1% of PTV volume received more than 108% of the prescribed dose. Institutional dose-volume constraints for OARs were as follows: Brainstem 1 cc <59 Gy, extended cord Dmax <45 Gy, parotid gland Dmean <25 Gy for node-negative neck site; V30 <50% for node-positive site, oral cavity beyond PTV <30 Gy. Image guidance was performed before each fraction; MVCT images of treatment area were taken, fused, and registered with both bony and tissue anatomy, whichever best fits PTVs and OARs. RT was delivered once daily, 5 days a week.

Patients except for T1N0 offered weekly 40 mg/m² cisplatin concomitantly and 3 cycles of adjuvant cisplatin and fluorouracil (CF) chemotherapy unless received induction chemotherapy (Carboplatin instead of cisplatin if there is renal impairment).

During RT, patients were monitored by the same treating physician (the same radiation oncologist did treatment plans and follow-ups during radiotherapy) for acute side effects, especially mucositis and dysphagia on weekly bases. Preliminary response assessment was done by Ear Nose Throat (ENT) specialist with a full head&neck examination, including fiberoptics imaging of nasopharynx (NP) one month after and with PET-CT 3 to 4 months after the end of RT unless suspect about disease dissemination. Follow-up was conducted every 3 months for the first 2-year and subsequently every 6 months for the following 3 years and yearly after that. During follow-up examinations, patients were evaluated for long-term side effects and any type of recurrence. Acute and late toxicities were scored according to the Radiation Therapy Oncology Group (RTOG) and the European Organization of Research and Treatment of Cancer (EORTC) toxicity criteria.[¹¹]

SPSS v20 was used for statistical analyses. Kaplan Meier method estimated loco-regional relapse-free survival (LRRFS), disease-free survival (DFS), distant metastasis-free survival (DMFS), and overall survival (OS). Loco-regional relapse-free survival was defined as the time between pathologic diagnosis of NPC and local or regional recurrence, death due to NPC, or unknown causes with the undocumented site of failure. Similarly, the event of interest was any type of recurrence (local, regional, distant) or death from any cause for DFS, distant metastasis or death for DMFS, death from any cause for OS. P value ≤0.05 was considered significant.

RESULTS

Pathologically confirmed staged 1-4 NPC patients treated in our HT unit were retrospectively reviewed. All patients except one completed the planned RT. Patients with distant metastasis at the time of diagnosis were excluded. The median age at diagnosis was 48.5 years (11–78) with male predominance (n = 54, 81.8%). The median age for female and male were 38 (IQR = 18) and 50 (IQR = 16), respectively. No significant difference was found between the median ages (Z = 1.317 P = 0.189).

Most common presenting symptoms were neck mass (n = 35; %52.3), hearing loss and ringing or fullness of ears (n = 16; %23.9) and nasal bleeding (n = 12; %17.9). Headache (n = 4; %4.5), blurry and double vision (n = 1; %1.5) were relatively uncommon symptoms. Pathological examination revealed 10.4% (n = 7) having keratinized squamous cell carcinoma (SCC), 32.8% (n = 22) non-keratinized SCC and 56.8% (n = 38) undifferentiated carcinoma. Patients and disease characteristics are shown in Table 1.

Patients were staged with PET-CT and MRI in 92.5% (n = 62); MRI and Thoracoabdominal CT in 7.4% (n = 5). According to TNM 8^th, 4.5% (n = 3) of the cases were stage 1, 16.4% (n = 11) were stage 2, 38.8% (n = 26) were stage 3 and 40.3% (n = 27) were stage 4.

All patients except one (T1N0) received concurrent cisplatin (40 mg/m2) or carboplatin (100 mg/m²) based weekly chemotherapy for median of 6 weeks (range, 3–7). Three patients referred from other facilities received 3 cycles of induction chemotherapy with TPF (cisplatin 75 mg/m² and docetaxel 75 mg/m² on day 1, 5-FU 750 mg/m² on day 1–4 every 3 weeks) and 44 (70%) patients received median 2 (range, 1–3) cycles of adjuvant PF (cisplatin 80 mg/m² on day 1 and 5-FU 800-1000 mg/m² on day 1–4 every 3 weeks).

Mean ± 11 Standard Deviation (sd) of D50 of patient's right and left parotid glands and cochleae were 33 ± 11 11 Gy; 31 ± 11 9 Gy and 37 ± 11 14 Gy; 34 ± 11 14 Gy, respectively. Mean ± 11 sd of Dmax of brainstem and medulla spinalis were 51 ± 11 11 Gy and 42 ± 11 12 Gy, respectively. These values were belonged to the first RT plan since all of the patients had at least one adaptive replan.

Patterns of failure and survival data

During the median 51 months (range, 2–100) follow up, 17 patients (29%) died, 12 of them related to disease progression, and toxicity (n = 8 and n = 4). Thirteen patients relapsed (19.4%), six of them (8.9%) recurred locoregionally, and 7 (10.4%) of them developed distant metastasis. The six patients with distant metastasis, 1 with local recurrence, and 1 with regional recurrence died. Acute neutropenia-related toxicity was the cause of death in 3 patients and late pharyngeal bleeding in one. One patient (T4N2) developed a second primary tumor at 32 months follow-up and died at 64 months. One patient (T4N2) had pathologically confirmed mediastinal lymph node recurrence (9^th month) had salvage mediastinal RT without systemic therapy, and he was alive at 25 months follow-up. Table 2 shows the outcome of the loco-regional recurrences. Locoregional relapse-free survival, DFS, DMFS and OS estimates at 2 and 5 years are 83% and 63%, 78.4%, and 61.7%, 83%, and 69%, 86%, and 71%, respectively [Figures 1 and 2].

Toxicity

Acute RT-related side effects were mostly grade 1 or 2 mucositis and dermatitis and resolved within months following RT and chemotherapy. Grade ≥3 dysphagia and mucositis observed in 28 (42%) patients required daily IV fluid transfusion and/or hyperalimentation on the last 2-3 weeks of RT. Three patients (4%) died within 3 months after diagnosis because of neutropenia-related infections (two of them were T3N2, and one of them died after the last weekly chemotherapy, and the other died after the 1^st cycle of adjuvant chemotherapy). One patient (T1N2) had grade 4 encephalitis, hospitalized but recovered and free of disease at 27 months follow-up. None of the patients had ≥grade 3 mucous membranes and salivary gland toxicity beyond 6 months. Two patients experienced ≥grade 3 late adverse events; one had grade 3 temporomandibular joint stiffness, and one died because of severe pharyngeal bleeding at 9 months follow up. One patient developed nasal synechia and had repeated surgical interventions.

DISCUSSION

In our study, 67 patients with NPC treated with SIB-IMRT with daily MVCT confirmation were reviewed. Patients received concurrent chemotherapy (n = 66) and either induction (n = 3) or adjuvant systemic therapy (n = 44). During the median 51 months follow-up, estimated 5-year LRRFS, DMFS, and OS rates were 85.7%, 85%, and 71%, respectively, with relatively low acute and late RT-related salivary gland toxicity rates.

Randomized clinical trials have shown benefit in terms of low toxicity and local control with IMRT over 2 and 3-dimensional techniques in locally advanced in NPC.[⁴⁵] HT is a unique way of IMRT and IGRT combination and can produce highly conformal dose distribution to the compound target volumes. Since it offers a steeper dose gradient, precise delineation, and treatment delivery is extremely important. MRI effectively shows the tumor extension of the structures around the base of the skull, such as brainstem/cranial nerves, and PET-CT helps to determine positive nodal volumes and distant metastasis more accurately.[¹²¹³] In the present study, the high rate of use of these modalities led to more proper contouring and staging, and the daily use of MVCT allowed more reliable RT delivery.

Despite advanced treatment techniques in NPC, isolated local and regional recurrences still occur either because of intrinsic radioresistance or insufficient dose nearby radiosensitive critical structures. The current treatment options include RT (Brachytherapy, Stereotactic radiosurgery, IMRT), surgery (Nasopharyngectomy, Neck Dissection), and chemotherapy either alone or in different combinations but, the upfront management of this type of failures remains to be determined. A significant proportion of recurrences can be successfully treated with 35%–70% 5-year local-regional control rates, whereas cranial involvement and N3 disease are not present.[¹⁴¹⁵] In our series, 4 of 6 loco-regional relapses were treated with ReRT and/or neck dissection, and obviously, the dosimetric advantage obtained by HT helped to make the optimal reirradiations possible.

Several randomized trials were established the survival advantage of combination RT with chemotherapy over RT alone in locally advanced NPC.[^16171819] The addition of chemotherapy also decreases local, regional, and distant recurrences. We found that except 4 patients, all received a median 6 cycles of weekly platin, and approximately 2/3 of the patients received adjuvant PF. This compares favorably to the compliance rates of another phase 3 randomized trials.[¹⁷] Although IMRT with chemotherapy achieves excellent loco-regional control rates, the development of distant organ metastasis still occurs and could not be salvaged and is the main cause of death.[⁸²⁰²¹] In the current study, all patients with DM (n = 7, 10.4%) received concurrent CRT, and 6 of them received additional adjuvant chemotherapy, but all of them except one died at the last follow-up, suggesting the need for more effective systemic therapies. Besides, neutropenia-related severe (≥3) systemic toxicity is a big concern; 3 patients died of sepsis, and one patient developed grade 4 encephalitis.

One of the main expected benefits of the IMRT planning technique is the prevention of excessive doses to the salivary glands, thus limits the severity of xerostomia. Mean ± 11 sd of D50 of parotid glands were 33 ± 11 11 Gy and 31 ± 11 9 Gy, which was quite bit more than other series,[²²²³] but none of our patients experienced severe xerostomia beyond 6 months post-RT. The degree of xerostomia is dependent largely on the dose/volume of the gland in the RT field, accordingly, adaptive IMRT is a good option to spare the parotid glands and limits xerostomia.[¹⁰] We believe that despite the high percentage of advanced-stage disease and high mean D50 of parotid glands, severe mucous membrane and salivary gland toxicities were low due to the adaptive IMRT re-plans.

The RTOG 0225 was the first multicenter SIB-IMRT feasibility study, which used a 70 Gy/33 regimen quite similar to the current study to be used as the standard RT of NPC.[⁸] They reported a 2-year LRFFS as 89.3% with an OS of 80.2%. Other studies documented more comparable results [Table 3], with an 83%-63% LRRFS and %86-71% OS at 2 and 5-year, respectively. The grade 5 toxic events (n = 3) occurred within the 1^st 3 months after NPC diagnosis may have led to low rates of LRRFS in our study. Moreover, about 10.5% and 32.8% of the patients in our study were WHO type 1 and type 2, and it is known that undifferentiated carcinomas have higher 5-year survival than those squamous cell carcinomas, which is not correct when considering 10-year survival.[²⁴] This can also explain our early-stage low survival rates.

SIB technique offers a relatively high fraction and prescription dose to the tumoral site, and combination with chemotherapy causes increased rates of serious adverse effects. In a study of Bakst et al. prescribed dose was 70 Gy/30 frx, and they reported varying degrees of temporal lobe necrosis in about 12% of patients.[²⁵] Du et al. reported their toxicity outcome in 2 different prescriptions (67.5 Gy/30 frx and 70-74/33 frx to primary tumor and nodes), and they found a 15.9% and 8.4% grade 3 non-hematologic acute toxicity, respectively.[⁹] Lee et al. documented grade 3 and 4 non-hematologic acute adverse effects in 72% of patients.[⁸] Toxicity rates were not correlated with each other in different series [Table 3]; the possible reason is the different prescriptions and evaluations by the doctors.

CONCLUSIONS

The present study reemphasized that adaptive SIB-IMRT with HT is a good option for the management of NPC with comparable loco-regional control rates and low salivary gland toxicity.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

REFERENCES