DOI: https://doi.org/10.21203/rs.3.rs-2456172/v1
Context: Toxicity from radiation therapy for laryngeal squamous cell carcinoma includes dysphagia, mucositis, laryngeal edema, weight loss and pain. Hypofractionation (HF) regimens shorten the overall treatment time to lower the risk for accelerated repopulation of tumor cells following initiation of radiotherapy (RT) demonstrated improved overall survival and locoregional control.
Objective: To examine the early toxicity of HF to the larynx, compared with standard fractionated (SF) RT.
Methods: Retrospective analysis of 127 laryngeal SCC patients who were treated with RT; Early glottic cancer (50%) received hypofractionation (2.25Gy/fr, 63Gy) only to the glottis while advanced stage disease (50%) received standard fractionation (2Gy/fr, 70Gy) to the glottis and to bilateral neck, with or without concurrent chemotherapy (CRT).
Results: Patients in HF group consumed significantly higher dosages of both Fentanyl (15 mg versus 8 mg) and Oxycodone (2.9 mg versus 2.1) due to increased pain and discomfort (P<0.05). Patients in HF group who were treated with Dexamethasone boluses demonstrated a significantly lower weight loss compared to HF patients not treated with steroids, and even promoted weight gain in some (p<0.005). Patients with advanced-stage cancer who were treated with CRT suffered from greater toxicity, compared to RT alone, leading to higher rates of dehydration events and weight loss.
Conclusions: HF-treated patients tend to consume significantly higher doses of opioids than patients treated with SF, representing higher levels of pain. Treatment regimens for patients undergoing HF should include proper pain management, and when possible, steroids.
Laryngeal cancer accounts for 4.5% of all malignancies and is the most common type of head and neck cancer [1], with approximately 13,000 new cases a year in the USA. The estimated overall 5-year survival rate for patients with glottic SCC is 60% [2], with early-stage (T1, T2) and advanced-stage (T3, T4) reaching 85%-95% [3, 4] and 50% [5], respectively. According to the American cancer society (ACS), all patients with early-stage disease should be treated with intent to preserve the larynx, either with surgery or with radiotherapy (RT) [6]. Nonsurgical larynx-preservation approaches for advanced-stage disease stem from the results of the Department of Veterans Affairs (VA) Laryngeal Cancer Study Group trial [7], and the Intergroup Radiation Therapy Oncology Group trial (RTOG 91 − 11) [8].
Radiotherapy techniques are aimed to induce tumor cell death by adhering to the R's of radiobiology and toxicity; i.e. repair of sublethal DNA damage, cell repopulation, redistribution of cells in the cell cycle, and reoxygenation of previously hypoxic tumor areas [9]. Toxicity to the larynx following RT is traditionally classified as early versus late phase and includes dysphagia and odynophagia (due to confluent mucositis), dyspnea, dysphonia, and recurrent aspirations, due to laryngeal edema and vocal cord dysfunction [10]. These in turn can cause weight loss, dehydration and in severe cases obstructive respiratory distress [11]. Toxicity greatly depends on the overall radiation dose and volume [12], and the addition of concurrent chemotherapy [13]. To lower the morbidity rates associated with RT, treatment plan includes fractionation of the total dose, and intensity-modulated technology (IMRT), to reduce damage to adjacent structures of the neck [14].
Fractionation of the total radiation dose allows cells that are in more radioresistant phases of the cell cycle to move to more sensitive stages and ultimately die. With standard fractionation (SF) radiotherapy, administered at 2Gy per fraction, most DNA damage does not result in immediate cell kill. Hypofractionation (HF) administers higher doses of radiation per dose (2.25Gy/fr), to increase tumor cell death, decrease repopulation of surviving clones and to improve overall survival. The cost for implementing HF radiotherapy is the risk for acute toxicity, resulting from increased damage to healthy cells, compared to SF [15]. HF radiotherapy is currently recommended for early glottic cancer, according to the recent United Kingdom national multidisciplinary guidelines [16].
In this study we aimed to examine the acute phase toxicity of HF radiotherapy to the larynx, as opposed to standard fractionated (SF) radiotherapy to larynx and neck, in patients who did not undergo surgery.
Following an institutional review board approval, all electronic medical charts of adult patients, over age 18 years, who were treated for laryngeal SCC at our institution during 2015–2018, were reviewed. Excluded were patients who underwent surgery or did not complete their treatment plan. The records of 162 patients were initially reviewed, and 127 patients were found eligible for the study criteria and were included in the study. All patients were staged by the 8th American Joint Committee on Cancer (AJCC) criteria. The patients were retrospectively assigned to 2 groups according to the RT fractionation regimen. Sixty-four patients with early glottic cancer received HF RT (2.25Gy/fr, 63Gy) to the glottis region versus sixty-three patients with advanced laryngeal cancer who were treated with SF RT (2Gy/fr, 70Gy) to the larynx and to the bilateral neck, with or without concurrent chemotherapy. The characteristics of patients and detailed features of treatment were shown in Table 1.
| Standard fractionation (SF) N = 63 | Hypofractionation (HF) N = 64 | Statistical significance P-value | |
|---|---|---|---|
| Number of patients (n) | 63 | 64 | 0.6 |
| Males (n) | 79% (50) | 85% (55) | 0.08 |
| Age (average ± standard deviation) | 63.9 ± 11.35 | 67.9 ± 10.4 | 0.04 |
| DM (n) | 24% (15) | 20% (13) | 0.6 |
| COPD (n) | 29% (18) | 9% (6) | 0.006 |
| Active smoking (n) | 46% (29) | 19% (12) | 0.001 |
| Past smoking (n) | 90% (57) | 70% (45) | 0.004 |
| Subsites | Glottic- 49%, Supraglottic- 46%, Subglottic-5% | Glottic- 100% | |
| Early versus Advanced Stage (n) | I/II 28% III/IV 72% | I/II 98% III/IV 2% | |
| T stage (n) | T1- 3% (2) T2-25% (16) T3-51% (32) T4-21% (13) | T1-76% (49) T2-22% (14) T3-2% (1) | |
| N stage (n) | N0- 72% (45) N1- 11% (7) N2- 14% (9) N3- 3% (2) | N0- 97% (62) N1- 3% (2) | |
| Metastases (n) | 13% (8) | 2% (1) | 0.009 |
| Recurrences (n) | 18% (11) | 2% (1) | 0.004 |
| Mortality (n) | 18% (11) | 8% (5) | 0.1 |
| Total dose (Gy) | 70 Gy (100%) | 63 Gy (72%) 65.2 Gy (18%) | 0.04 |
| Concurrent chemotherapy (n) | 79% (50) | 0 | 0.001 |
| DM; Diabetes Mellitus, COPD; Chronic Obstructive Pulmonary Disease | |||
We hypothesized that HF in laryngeal cancer patients causes higher rates of acute toxicity, due to higher cell damage to healthy cells, compared with SF regimens. To assess toxicity, we looked at different outcomes: consumption of opioids (as proxy to pain assessment), weight loss, mucositis, laryngeal edema, dehydration events, treatment breaks and hospitalization rates.
Dosimetric measurements were performed according to ICRU recommendations [17]. Radiotherapy was delivered by LINAC to patient placed in supine position, with head fixed in a thermoplastic mask. CT imaging was used for multi-slice IMRT planning with VARIAN. Patients with T1 and T2 glottic cancer, were treated for a total dose of 63 and 65.25 Gy, respectively, in 2.25Gy fractions delivered only to the larynx (HF). Patients with T3 N0 glottic cancer and T1–3 N0-3 supraglottic cancer were irradiated with 35 fractions of 2Gy to larynx and gross lymphatic neck spread and 60Gy in 1.8Gy fractions to elective volumes (bilateral levels II–VI). Table 1 presents fractionation in study and control groups.
The tumor biologically effective dose (BED) was calculated using the standard linear quadratic equation [18]; BED = nd(1 + d)/[α/β])-loge2 (T-Tk)/ αTp.
D = total dose (Gy), d = dose per fraction (Gy), α/β = linear (α) and quadratic (β) components of the linear quadratic model (Gy); T = overall treatment time (days); Tk = onset of accelerated repopulation time (days); Tp = average doubling time during accelerated repopulation (days).
All statistical analyses were performed using IBM SPSS Statistics for Windows, Version 27.0 (IBM Corp, Armonk, NY, USA). A p-value cut-off point of 0.05 at 95% confidence interval (CI) was used to determine statistical significance. Comparison of categorical variables between the two independent groups was conducted using the Pearson chi-square test, and logistic regression analysis, while t-test and ANOVA were used to compare normally distributed continuous variable. Post-hoc multiple logistic regression analysis was used to assess the outcome measures according to the risk factors.
Overall, 127 patients were included in the research, categorized into the SF and HF groups, comprising of 63 and 64 patients, respectively. Average age (years) in SF was 63.9 ± 11.35, which was younger compared to 67.9 ± 10.4 in HF (p = 0.04). Other statistically significant differences (p < 0.05) were seen; SF patients had higher rates of COPD (29%) compared to HF patients (9%), higher rates of past smoking (90% vs 70%) and more active smoking (46% vs 19%). General characteristics and medical history comparison is depicted in Table 1.
Tumor subsites were glottic (49%), supraglottic (46%) and subglottic (5%) in the SF group, while there were exclusively glottic tumors (100%) in the HF group (p = 0.001). In the SF group, 29% of the patients and 71% of the patients had early (I/II) and advanced stage (III/IV) laryngeal tumors, respectively, compared to 98% and 2% in the HF group who presented with early and advanced-stage disease, respectively (p = < 0.001). Detailed tumor staging between the two groups, as well as recurrence rates and all-cause mortality are depicted in Table 1.
Prescription and usage of opioids during radiation treatments were evaluated and compared between the SF and HF groups. In SF 35% patients were prescribed Fentanyl patches, compared to 44% in HF group (p = 0.2). The mean dosage of Fentanyl patches in HF patients was 15 mg for every 72 hours, compared to 8.4 mg in SF patients (p = 0.002). Similarly, the mean SOS dosage of Oxycodone was also higher in HF patients, compared to SF patients (3.2 mg versus 2.1 mg, in HF and SF, respectively), although not statistically significant (p = 0.08); Moreover, higher rates of opioid use for longer periods (more than one month), was seen in HF (24%) compared to SF (14%), although not statistically significant (p = 0.2). Results are depicted in detail in Table 3. A sub-group comparison analyses based on smoking status and steroidal treatment during radiation therapy, failed to find any significant differences in Fentanyl and Oxycodone dosages for these factors in either the SF or HF groups (data not shown). A comparison within the HF group, between T1 patients treated with 63Gy and T2 patients treated with 65.5Gy also did not find any significant differences regarding Fentanyl or Oxycodone dosage (data not shown).
Weight loss rates (defined as loss of more than 1 kg) were comparable and seen in 51 (81%) and 48 (75%) patients in the SF and HF groups, respectively (N.S); with an average weight loss of 4.2 and 3.2 kg in SF and HF, accordingly (N.S). Steroids were given to 31% of patients, in both groups. Patients in HF group who received steroids demonstrated a mild weight gain (1.3kg, p = 0.001), compared to HF patients who did not receive steroids. Further details are depicted in Table 2 and Fig. 1.
|
Standard fractionation (SF), N = 63 |
Hypofractionation (HF), N = 64 |
Statistical significance P-value |
|
|---|---|---|---|
|
Hospitalization |
19% (12) |
13% (8) |
0.3 |
|
Oral mucositis |
13% (8) |
14% (9) |
0.8 |
|
Laryngeal mucositis |
8% (5) |
8% (5) |
0.95 |
|
Treatment break |
13% (8) |
9% (6) |
0.4 |
|
Dehydration |
48% (30) |
23% (15) |
0.03 |
|
Steroids |
31% (20) |
31% (20) |
0.95 |
|
Opioids during treatment |
35% (22) |
45% (29) |
0.2 |
|
Opioids 1 month after end of treatment |
14% (9) |
24% (15) |
0.2 |
|
Average Fentanyl patch dosage |
8.4 ± 16.5 |
15.7 ± 22.8 |
0.002 |
|
Average SOS Oxycodone dosage |
2.1 ± 3.1 |
2.9 ± 3.6 |
0.08 |
|
Average weight loss (Kg) |
-4.2 ± 4.6 |
-3.2 ± 7.3 |
0.6 |
In a multivariate logistic regression analysis, comparing odds ratio in HF versus SF patients, the following variables were found significant: Male gender (OR = 2.4, p = 0.002), age over 60 (OR = 3.1, p = 0.003), Max Fentanyl dose (OR = 1.2, p = 0.03), average weight loss more than 1 kg (OR = 0.14, p = 0.002).
In subgroup analysis of SF patients, a trend of increased toxicity was seen in those who were given CT compared to RT alone. Rates of hospitalization, dehydration, laryngeal and oral mucositis were higher, although not significant, in patients undergoing concomitant chemotherapy with RT. Correspondingly, there was a significant increase in the use of steroids and opioids (both fentanyl patch and Oxycodone) in SF patients undergoing chemoradiotherapy compared with RT alone (Table 3 and Fig. 1).
|
Standard fractionation RT (N = 13) |
Standard fractionation RT + CT (N = 50) |
Statistical significance P-value |
|
|---|---|---|---|
|
Average Fentanyl patch dosage |
2.9 ± 4.3 |
9.8 ± 5.1 |
0.05 |
|
Average SOS Oxycodone dosage |
0.9 ± 1.5 |
2.4 ± 0.9 |
0.002 |
|
Hospitalization |
15% (2) |
20% (10) |
0.8 |
|
Oral mucositis |
8% (1) |
16% (8) |
0.4 |
|
Laryngeal mucositis |
8% (1) |
8% (4) |
0.6 |
|
Dehydration |
33% (5) |
54% (27) |
0.1 |
|
Steroids |
15% (2) |
36% (18) |
0.03 |
|
Average weight loss (Kg) |
-2.8 ± 3.69 |
-4.6 ± 4.8 |
0.6 |
In early glottic cancer, several altered fractionation regimes that shorten the overall treatment time, have proven beneficial for long term outcomes, as they lower the risk for repopulation [19]. Fractionation enables the administration of higher total radiation doses that are tolerated by the patients. While long term overall survival and disease-free survival favor the use of HF RT in early glottic cancer patients, the treatment-related morbidity and laryngeal toxicity are often overlooked.
In head and neck cancer patients, pain is a common symptom in 50% of patients prior to treatment (tumor-related pain). During radiotherapy more than 80% of patients report pain, which is considered treatment-related toxicity. Moreover, in a third of head and neck cancer patients, the pain continues for more than 6 months after cessation of treatment [20]. This well-known treatment-related pain is considered both nociceptive and neuropathic in nature [21]. Side effects from radiotherapy, specifically those occurring during the treatment period, (i.e., acute toxicity), are dose dependent. Hypofractionated radiotherapy administers higher radiation doses per fraction to patients and was shown to cause higher rates of acute toxicity in other cancers, outside the head and neck area [22]. Opioids are a cornerstone in the treatment of radiation-induced pain in head and neck cancer patients. Opioid dosages are often used as indicators of pain levels in these patients, as they are often correlate [23].
Our study is the first to report on pain levels in HF compared to SF patients. We compared laryngeal toxicity among 127 patients who were treated with either SF or HF RT. Interestingly both Oxycodone and Fentanyl dosages were higher, and for longer periods of time, in HF groups, compared to SF patients. Although a direct measure of pain was not performed routinely in this research (via for instance the Visual Analogue Scale; VAS), given the reasons mentioned here, the authors believe that the increase in opioid dosage seen in hypofractionated patients signifies the increase in pain levels associated with the treatment plan. Interestingly, pain and discomfort were significantly higher in HF group, even when compared to SF who also received CRT. Furthermore, although T2 patients were treated with a higher overall radiation dose (65.5Gy) for longer periods than T1 patients, their fraction dose remained the same (2.25Gy per dose). No difference was seen in terms of Fentanyl and Oxycodone doses between these subgroups of HF patients. These results further strengthen the importance of fraction dose over the total dose on early RT related toxicity.
Our results are unique since HF is not usually associated with aggravation of pain. Other altered fractionation regimens are often reported to cause laryngeal toxicity; Hliniak et al. [24], demonstrated higher rates of early toxicity including pain for patients following Hyperfractionation to the larynx (66Gy in 33 fr; BID on Thursdays). Trotti et al. [25], reported early skin, mucosal, and laryngeal toxicity in hyperfractionating regimen (1.2 Gy BID; 79.2 Gy/66 fr) compared to SF. Higher rates of early toxicity were also reported in accelerated regimen published by The Danish Head and Neck Cancer Group [26].
Contrary to these altered regimens, pain level is not routinely reported in studies that compare toxicities in hypofractionation (HF) RT for glottic SCC [27, 28]. The limited publications that addressed pain in HF patients report similar levels compared to SF. Motegi et al. [29] assessed acute toxicity among 148 patients with T1-2 glottic SCC, treated with HF RT (2.4Gy), and reported that none of the patients had experienced pain following treatment. Kodaira et al. [30] found similar levels of pain in patients treated with either SF or HF for early glottic SCC.
Higher rates of Opioid consumption were previously reported in a single publication by Scott et al. [31]. In their study, 52 patients received 2.25Gy per fraction for early glottic SCC, and consumed more narcotics than 48 patients that were treated with SF. Our results support the study by Scott et al, and further emphasize the importance of pain management in this subgroup of relatively young and healthy patients undergoing hypofractionated radiotherapy (2.25Gy). Radiation-induced severe pain (requiring opioids), is associated with higher rates of premature discontinuation of curative radiation therapy (PDCRT), as seen in a recent publication by Lazarev et al. [32]. In their study, 67% of patients experienced significant pains, ultimately leading to PDCRT. Treating physicians should therefore anticipate higher levels of pain among HF patients, educate them in advance and encourage them to report and disclose pains during early stages of treatment. By providing appropriate analgesics, we facilitate treatment and improve patient compliance.
Weight loss during oncological treatment is an important consideration in head and neck cancer patients, as it affects quality of life, mortality rates and treatment outcomes [33]. Many factors contribute to the dysphagia and associated weight loss seen in non-surgical patients treated for laryngeal SCC, such as laryngeal edema [34], chemotherapy induced mucositis [35], or suppression of appetite due to Opioid use. Steroids are sometimes given during treatment to increase appetite and to treat anorexia in palliative patients [36]. They also aid to augment well-being in oncologic patients [37].
In our study, average weight loss (kg) was similar among SF and HF patients, as well as similar rates of weight loss over 1 kg. Interestingly, although a third of the patients in both SF and HF groups received Dexamethasone to treat cachexia, severe pain and to improve overall well-being, only HF patients that were treated with steroids demonstrated lower weight loss and even mild weight gain, compared to HF that were not given steroids. These beneficiary effects of steroidal treatment were not seen in SF patients. These results demonstrate the importance of introducing steroids to hypofractionated RT regimens to aid with dysphagia and weight loss, as opposed to standard fractionation. It should be noted that Dexamethasone did not lower opioid use or dosage in either HF or SF patients.
The morbidity related to the aggressive treatment in SF group, and the overall impaired physical condition due to pre-existing co-morbidities led to significantly higher rates of dehydration events and weight loss, compared to HF patients. Significant differences in toxicity were seen among patients in SF group who received CRT compared to RT alone. Most of the dehydration events in SF group was related to the CT. Furthermore, significantly higher opioid dosages were seen in CRT patients, compared to RT patients in SF group. Interestingly, Fentanyl dosages were higher in early glottic cancer patients receiving only HF RT compared to advanced-stage patients treated with both radiotherapy and chemotherapy.
Contrary to HF, patients in SF were already given dexamethasone as part of Cisplatin regimen. The additional steroids given to a third of SF patients (those who were also treated with CT) did not prevent weight loss, nor promote weight-gain in these patients. These results demonstrate the laryngeal and overall morbidity associated with chemotherapy, as demonstrated in other studies [38].
The main limitation of our study is the retrospective analysis, and the lack of randomization to the treatment arms. Our patients were treated according to the standard of care within each cancer group according to stage. Therefore, we were unable to compare toxicity of HF and SF RT within same-stage cancer patients. Another limitation is the assessment of pain levels via opioids dosages and not via designated questionnaires.
Hypofractionation is not without cost. Patients treated with HF tend to consume significantly higher doses of opioids than patients treated with SF, representing higher levels of pain. Pain in early phase toxicity is affected more by fractionation dose than overall dose and treatment area. Pain affects quality of life and patient compliance to therapy and should there be addressed during treatment planning and clinic follow ups. We encourage physicians who treat glottic SCC patients with HF RT to consider incorporating both opioids and steroids into the treatment regimens. Future studies are needed to assess pain levels directly (via designated questionnaires for instance).
Ethical Approval
The study was approved by the institutional IRB no. RMC-2018-342.
Competing interests
Authors' contributions
O.Z. designed the work, analyzed and interpreted the results, and wrote the final draft; E.Y. analyzed and interpreted the results; H.S.H. and Y.H. revised the draft and approved the final version; O.I. and D.S.Y. acquired the results; G.B. revised the manuscript; N.G. designed the work, interpreted the results, and approved the final version.
Funding
Availability of data and materials
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Submission Declaration