Abstract
In locally advanced head and neck squamous-cell carcinoma (LA-HNSCC), clinical complete response (cCR) at the primary site, assessed by clinical examination, after induction chemotherapy predicts for a low relapse risk after subsequent chemoradiotherapy. Prior studies showed a cCR rate of 77% with induction nanoparticle albumin-bound (nab)-paclitaxel given with cisplatin and 5-fluorouracil (APF). The primary aims of this non-randomized phase 2 trial were to determine the cCR rate after induction nab-paclitaxel and cisplatin (Arm 1) and after nab-paclitaxel monotherapy (Arm 2). Eligibility required LA-HNSCC, T2-T4 stage classification, and suitable (Arm 1) or unsuitable (Arm 2) candidates for cisplatin. Arm 1 patients received nab-paclitaxel and cisplatin, then cisplatin with radiation. Arm 2 patients received nab-paclitaxel, then cetuximab with radiation. The primary endpoint was cCR after two cycles of induction chemotherapy. Each arm enrolled forty patients. cCR at the primary site occurred in 28 patients (70%) after nab-paclitaxel and cisplatin and in 8 patients (20%) after nab-paclitaxel monotherapy. The overall clinical response rate was 98% after nab-paclitaxel and cisplatin and 90% after nab-paclitaxel monotherapy. In subset analyses, cCR rates by T stage classifications (T2, T3, T4) were 54, 86, and 69% after nab-paclitaxel and cisplatin, and 14, 11, and 26% after nab-paclitaxel. cCR rates by human papillomavirus status (p16 positive oropharynx vs other) were 72 and 64% after nab-paclitaxel and cisplatin and 35 and 9% after nab-paclitaxel. The cCR rate after nab-paclitaxel and cisplatin was similar to APF; however, the cCR rate after nab-paclitaxel monotherapy was lower. The trial was registered at ClinicalTrials.gov NCT02573493 on October 9, 2015.
Keywords: nab-paclitaxel, Head and neck squamous-cell carcinoma, Cisplatin, Cetuximab, Radiation
Introduction
Patients with locally advanced head and neck squamous-cell carcinoma (LA-HNSCC) are often treated with concurrent chemoradiotherapy (CRT) [1]. The “gold standard” CRT regimen is cisplatin with radiation therapy (CisRT) [1]. Suitable candidates for CisRT have good performance status and organ function. However, many patients do not meet these criteria [2]. In cisplatin-unsuitable patients, there are no completed randomized trials to guide treatment selection. Observational studies showed that cetuximab was the most common drug combined with radiation (CetuxRT) in patients not given cisplatin [3].
Relapse of disease is the dominant cause of treatment failure after CRT. The risk of relapse after CRT directly correlated with pre-treatment tumor volume [4]. Induction paclitaxel or docetaxel given with cisplatin and 5-fluorouracil reduced tumor volume before CRT, but randomized trials in non-nasopharyngeal cancer showed inconsistent effects on relapse risk and no overall survival (OS) benefit [5-8]. Prior studies showed that clinical complete response (cCR) at the primary site, assessed by clinical examination, after induction chemotherapy predicted for a low risk of relapse after subsequent CRT; however, the rates of cCR after current induction regimens were low. Strategies to increase the cCR rate after induction therapy may reduce the relapse rate after CRT.
Macropinocytosis is an endocytotic, nutrient-scavenging pathway that promotes internalization of albumin into cells [9]. Albumin serves as a major source of nutrients and energy. Macropinocytosis is driven by activation of the EGFR, RAS, and PIK3CA signaling pathways [10, 11]. These pathways are often constitutively activated in HNSCC [12]. In this way, anti-tumor drugs bound to albumin may selectively target cancer cells with upregulation of macropinocytosis. nab-paclitaxel is a nanoparticle albumin-bound formulation of paclitaxel. nab-paclitaxel and paclitaxel are not equivalent drugs. In pre-clinical models, drug delivery into cancer cells and anti-tumor activity were better with nab-paclitaxel compared to paclitaxel [13]. In advanced squamous-cell lung and breast cancers, tumor response rates were higher with nab-paclitaxel compared to paclitaxel [14, 15]. In addition, nab-paclitaxel is active in pancreatic cancer and melanoma, whereas, paclitaxel is not [16, 17].
We reported two consecutive, single-arm, phase 2 trials that incorporated nab-paclitaxel into cisplatin-based induction regimens to achieve tumor volume reduction before CisRT [18, 19]. The first trial’s regimen was nab-paclitaxel, cisplatin, 5-fluorouracil, and cetuximab (APF-C) [18]. The regimen for the second trial was APF [19]. Patients with favorable tumor response (complete or partial) at the primary site, assessed by clinical examination, after two cycles of induction therapy received one additional cycle followed by CisRT. Patients without favorable tumor response proceeded directly to CisRT. In both trials, the favorable tumor response rates after two cycles of induction therapy were ≥ 94%, and after completion of CisRT, the local–regional control (LRC) rates were ≥ 90% and relapse rates were ≤ 17% [18-20]. The cCR rate at the primary site after two cycles of APF was 77%. No clear signal of benefit from cetuximab was observed. Other investigators reproduced these high rates of tumor response and LRC with nab-paclitaxel-based induction regimens [21, 22].
To assess the contribution of 5-fluorouracil and of cisplatin given in combination with nab-paclitaxel to achieve tumor volume reduction before CRT, we performed a multicenter, non-randomized phase 2 trial. In Arm 1, cisplatin-suitable candidates received nab-paclitaxel and cisplatin followed by CisRT. In Arm 2, cisplatin-unsuitable candidates received nab-paclitaxel monotherapy followed by CetuxRT. We hypothesized that each induction regimen would result in a cCR rate at the primary site similar to our historical experience with APF [19].
Materials and methods
Patient population
Eligible patients had untreated LA-HNSCC (stage III or IV, American Joint Committee on Cancer, 7th edition) that originated in the oropharynx (OP), larynx, or hypopharynx, T2-T4 primary site stage classification, and measurable disease per Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST1.1) [23]. Patients were 18 years or older with adequate blood counts [absolute neutrophil count (ANC) ≥ 1500/μL, platelet count ≥ 100,000/mL, hemoglobin ≥ 9 g/dL]. One of the following criteria was required: candidate for organ preservation, unresectable disease, unacceptable morbidity of surgery, or patient declined surgery. Patients on Arm 1 were required to meet all criteria for cisplatin: ECOG performance status 0–1 and adequate organ function [serum creatinine < 1.5 mg/dL or GFR ≥ 75 cc/min; bilirubin < 1.6 mg/dL; AST and ALT < 2.5 × upper limits of normal (ULN)]. Patients on Arm 2 were required to have one or more criteria of unsuitability for cisplatin: ECOG performance status 2, reduced GFR (31–74 cc/min), moderate/severe chronic obstructive pulmonary disease, profound hearing loss, or organ transplant. Key exclusion criteria included T1 stage classification and ≥ Grade 2 peripheral neuropathy. The institutional review board at each of the four participating sites approved the protocol. All study participants provided written, signed, informed consent to participate. The quality assurance committee of Washington University (St. Louis, MO, USA) performed independent data monitoring.
Investigational treatment
Before treatment, patients underwent physical examination, laboratory evaluation, electrocardiogram, pulmonary function tests, computed tomography (CT) scans of the neck and chest, and 18fluorodeoxyglucose (FDG)-positron emission tomography (PET)/CT imaging. Immunohistochemistry was performed on OPSCC archival tumor tissue to assess expression of p16INK4A, a surrogate marker of the human papillomavirus (HPV) [19]. Patients on both arms were treated with nab-paclitaxel 100 mg/m2 administered intravenously (IV) on days 1, 8, and 15 of each 3 week cycle. Patients on Arm 1 also received cisplatin 75 mg/m2 IV administered on day 1 of each cycle.
After two cycles of induction therapy, patients underwent an assessment of tumor response at the primary site by clinical examination, CT of the neck, and FDG-PET/CT. The same otolaryngology physician who performed the initial evaluation also performed endoscopic and/or clinical examinations to determine response at the primary site, as previously described [19]. Clinical tumor response categories included cCR—no residual tumor by examination, clinical partial response (cPR)—at least 50% reduction in the longest dimension of the lesion, clinical stable disease (cSD)—less than 50% reduction but no progression, and clinical progressive disease (cPD)—interval enlargement. Patients with cCR or cPR at the primary site received a third cycle of induction therapy prior to CRT, whereas patients with cSD or cPD proceeded directly to CRT.
CRT began 1–4 weeks after completion of induction therapy. In Arm 1, CisRT was administered if the GFR was ≥ 60 cc/min; if not, CetuxRT was administered. CisRT included cisplatin 100 mg/m2 on days 1, 22, and 43 of RT. In Arm 2, CetuxRT was administered. Cetuximab 400 mg/m2 IV loading dose was administered one week before RT, then 250 mg/m2 weekly × 7 concurrent with RT. Intensity-modulated radiation therapy (IMRT) was delivered as follows: 7000 cGy in 35 fractions to gross disease (PTV1), an optional dose of 6300 cGy in 35 fractions to intermediate risk areas (PTV2), and 5600 cGy in 35 fractions (PTV3) to regions in the ipsilateral and contralateral neck at risk for microscopic disease. After completion of therapy, patients underwent physical examination and imaging every 4–6 months.
Endpoints and statistical methods
The primary endpoint for each arm was the proportion of patients achieving a cCR at the primary tumor site after two cycles of induction therapy. For each arm, we hypothesized that the cCR rate after two cycles of induction therapy would be similar to APF (77%) [19]. As the cCR rate with APF-C was 53% and the LRC rates with both regimens were high (≥ 90%) [18-20], we choose to set the lower boundary for a non-clinically significant difference for the cCR rate at 58%. Forty patients in each arm provided a power = 0.80 with a one-sided type I error of 0.05 to conclude that nab-paclitaxel and cisplatin or nab-paclitaxel monotherapy were not clinically different from the historical reference. Subset analyses were performed to determine the cCR rate at the primary site for each T classification and in HPV-positive and HPV-negative disease. HPV-positive disease was defined as p16 positive OPSCC and HPV-negative disease was p16-negative OPSCC, or SCC of the larynx or hypopharynx. In each arm, we monitored the rate of unfavorable clinical response (cSD and cPD) at the primary tumor site after two cycles of induction therapy using a continuous efficacy-monitoring rule. The historical unfavorable clinical response rate with APF was 7% [19]. The maximum allowable unfavorable clinical response rate at the primary tumor site for each arm was 20%. Enrollment in each arm would be suspended for excess unfavorable responses at the primary tumor site.
Secondary endpoints were the proportion of patients in each arm achieving a cPR, cSD, or cPD at the primary site and a cCR, cPR, cSD, or cPD at the neck nodes after two cycles of induction therapy; tumor response by RECIST1.1 and PERCIST [24] after two cycles of induction therapy; AEs; progression-free survival (PFS); and OS.
AEs were graded using NCI-CTCAE version 3.0. PFS was the time from diagnosis to disease progression, death, or last follow-up alive. The alive patients without progression were censored at the date of the last follow-up. OS was the time from diagnosis to death or last follow-up alive. The alive patients were censored at the date of the last follow-up. LRC was the time from diagnosis to local–regional relapse or last follow-up alive. The alive patients without local–regional relapse were censored at the date of the last follow-up. Survival endpoints and LRC were estimated by the Kaplan–Meier (KM) method [with 95% confidence intervals (CI)]. Follow-up was defined as time from diagnosis to last date alive, date of withdrawal or death. RT delivery (number of fractions, elapsed days, total dose) and chemotherapy administration (number of planned doses administered) were summarized using proportions, with medians and ranges. Proportions of patients with cPR at the primary tumor site, cCR and cPR rates at the involved regional nodes, anatomic and metabolic tumor responses, and AEs by type, grade, and patient were calculated.
The proportion of patients who experienced grade 3–4 AEs during three cycles of APF was 40% [19]. As a secondary endpoint, we hypothesized that the proportion of patients who experienced grade 3–4 AEs would be at least 25% lower with nab-paclitaxel and cisplatin and 50% lower with nab-paclitaxel monotherapy, in comparison with historical (APF). One-sided tests for differences of binomial proportions were used to test for differences in AE rates with a false discovery rate adjustment for multiple testing.
Results
Eighty patients enrolled between 2016 and 2019: 40 patients to each arm (Table 1). Most patients had a smoking history, T3 or T4 stage classification, and stage IV disease. In comparison to the historical reference (APF), patients in Arm 1 had similar patient and tumor characteristics, whereas patients in Arm 2 were older and had poorer performance status and higher T stage classification. In Arm 2, the most common reasons for cisplatin-unsuitability included reduced creatinine clearance, ECOG performance status 2, and profound hearing loss. The trial profile is shown in Fig. 1. Two cycles of induction therapy were administered to 79 patients (99%), and three cycles were administered to 74 patients (93%). In Arm 1, 34 patients (85%) received CisRT. In Arm 2, 35 patients (88%) received CetuxRT. Dose modifications and treatment delivery are shown in table 1 of the appendix.
Table 1.
Patient and tumor characteristics in Arms 1 (nab-Paclitaxel + Cisplatin) and 2 (nab-Paclitaxel), compared to the historical induction regimen (APF)
| Characteristic | Arm 1 (n = 40) | Arm 2 (n = 40) | APFa (n = 30) | p value | |
|---|---|---|---|---|---|
| (Arm 1 vs APF) | (Arm 2 vs APF) | ||||
| Age (years) | 0.47 | 0.003 | |||
| Median | 58 | 66 | 58 | ||
| Range | 42–78 | 49–84 | 43–75 | ||
| Gender (#) | 0.48 | 0.72 | |||
| Male | 36 | 32 | 25 | ||
| Female | 4 | 8 | 5 | ||
| Smoker (#) | 0.40 | 0.74 | |||
| Yes | 27 | 32 | 23 | ||
| ≥ 10 pack years | 25 | 30 | 20 | ||
| < 10 pack years | 2 | 2 | 3 | ||
| No | 13 | 8 | 7 | ||
| ECOG performance status (#) | 0.93 | 0.002 | |||
| 0 | 31 | 16 | 23 | ||
| 1 | 9 | 13 | 7 | ||
| 2 | 0 | 11 | 0 | ||
| Primary site (#) | 0.31 | 0.93 | |||
| Oropharynx | 31 | 26 | 18 | ||
| Larynx | 7 | 10 | 9 | ||
| Hypopharynx | 2 | 4 | 3 | ||
| p16 status (#) | 0.17 | 0.24 | |||
| HPV+OPSCC | 29 | 17 | 17 | ||
| HPV−HNSCC | 11 | 23 | 13 | ||
| T classification (#)b | 0.76 | 0.04 | |||
| 2 | 13 | 7 | 8 | ||
| 3 | 14 | 9 | 13 | ||
| 4 | 13 | 24 | 9 | ||
| N classification (#)b | 0.62 | 0.08 | |||
| 0-2a/b | 17 | 23 | 11 | ||
| 2c-3 | 23 | 17 | 19 | ||
| Stage (#)b | 1.00 | 0.72 | |||
| III | 4 | 6 | 3 | ||
| IV (a/b) | 36 | 34 | 27 | ||
| Reason for non-surgical therapy (#) | – | – | |||
| Unacceptable morbidity of surgery | 26 | 21 | – | ||
| Organ preservation | 8 | 13 | |||
| Unresectable | 3 | 5 | |||
| Declined surgery | 3 | 1 | |||
| Reason cisplatin-unsuitable (#) | N/A | N/A | |||
| Creatinine clearance (30-74 cc/min) | N/A | 13 | N/A | ||
| ECOG 2 | 9 | ||||
| Profound hearing loss | 9 | ||||
| Poor pulmonary function | 6 | ||||
| Organ transplant | 2 | ||||
| Other—cardiac | 1 | ||||
Fig. 1.
Trial profile
Primary endpoint: cCR at the primary tumor site after induction therapy
In Arm 1, 28 patients (70%) achieved cCR at the primary site after two cycles of nab-paclitaxel and cisplatin. In Arm 2, 8 patients (20%) achieved cCR after nab-paclitaxel monotherapy.
The proportions of patients in Arm 1 with cCR at the primary site after two cycles of induction therapy were 54% for T2 disease, 86% for T3, and 69% for T4. The proportions of patients in Arm 2 with cCR were 14% for T2 disease, 11% for T3, and 26% for T4 (Table 2). The proportions of patients in Arm 1 with cCR at the primary site were 72% for HPV-positive OPSCC and 64% for HPV-negative HNSCC. In Arm 2, the proportion of patients with cCR were 35% for HPV-positive OPSCC and 9% for HPV-negative HNSCC (Table 2).
Table 2.
Clinical Tumor Response at the Primary Site, Stratified by T Stage Classification and Human Papillomavirus (HPV) Status
| Stratification factor | Clinical tumor response at the primary site | |||||
|---|---|---|---|---|---|---|
| Complete | Partial | Stable | ||||
|
|
|
|
||||
| Arm 1 | Arm 2 | Arm 1 | Arm 2 | Arm 1 | Arm 2 | |
| All patients | 28 | 8 | 11 | 28 | 1 | 3 |
| T classification | ||||||
| T2 | 7 | 1 | 6 | 6 | 0 | 0 |
| T3 | 12 | 1 | 1 | 6 | 1 | 2 |
| T4 | 9 | 6 | 4 | 16 | 0 | 1 |
| HPV status | ||||||
| Positive | 21 | 6 | 8 | 9 | 0 | 2 |
| Negative | 7 | 2 | 3 | 19 | 1 | 1 |
Secondary endpoint: tumor response after induction therapy
The favorable (cCR + cPR) clinical tumor response rates at the primary site after two cycles of induction therapy were 98% in Arm 1 and 90% in Arm 2 (Table 2).
After two cycles of induction therapy, the proportions of patients who achieved an overall (primary site and neck nodes) favorable tumor response by clinical examination were 95% in Arm 1 and 83% in Arm 2 (Table 2). The proportions of patients who achieved an overall tumor response by RECIST were 60% in Arm 1 and 53% in Arm 2 (Table 2 of the appendix). The proportions of patients who achieved an overall tumor response by PERCIST were 95% in Arm 1 and 67% in Arm 2 (Table 2 of the appendix). No patient in either arm had progressive disease at the primary site during induction therapy. Two of eighty patients (both in Arm 2) had progressive disease during induction therapy.
Secondary endpoints: relapse, LRC, and survival
In Arm 1, the median follow-up of the alive patients was 34 months (IQR 27–40). Relapse occurred in 5 patients (12.5%), and death occurred in 5 (4 relapse, 1 s cancer). Among the 29 patients with HPV-positive OPSCC, the LRC, PFS, and OS at 2 years were 100% (95%CI 100–100), 90% (95% CI 71–97), and 100% (95% CI 100–100), respectively (Fig. 2). Among the 11 patients with HPV-negative HNSCC, the LRC, PFS, and OS at two years were 89% (95% CI 43–98), 82% (95% CI 45–95), and 89% (95% CI 43–98), respectively.
Fig. 2.
Local–regional control (LRC) and survival curves for Arms 1 and 2, stratified by human papillomavirus (HPV) status. a Local–regional control, b progression-free survival, c overall survival
In Arm 2, the median follow-up of the alive patients was 20 months (IQR 15–30). Relapse occurred in 13 patients (32.5%), and death occurred in 12 (7—relapse, 1—co-morbidity, 2—unknown, 1—sepsis, 1—suicide). Among the 23 patients with HPV-negative HNSCC, the LRC, PFS, and OS at 2 years were 72% (95% CI 45–88), 54% (95% CI 32–72), and 67% (95% CI 43–83), respectively (Fig. 2). Among the 17 patients with HPV-positive OPSCC, the LRC, PFS, and OS at 2 years were 85% (95% CI 51–96), PFS 53% (95% CI 25–75), and OS 71% (95% CI 38–88), respectively. Of these 17 patients, 16 had one or more high-risk prognostic characteristics (T4, N2c or N3 classification, or ≥ 10 pack years of smoking).
Secondary endpoint: toxicity
One or more grade 3–4 AEs occurred during induction therapy in 23 patients (58%) in Arm 1 and 17 patients (43%) in Arm 2 (Table 3). The proportion of patients in each arm with selected grade 3–4 gastrointestinal (diarrhea and/or oral ulcers) toxicities that occurred during induction therapy was 3%. One treatment-related mortality occurred, a patient on Arm 2 who experienced an intravenous catheter-related blood infection with normal neutrophil counts. One or more grade 3–4 AEs occurred during CRT in 37 of 39 patients (95%) in Arm 1 and in 25 of 35 patients (71%) in Arm 2 (Table 3 of the appendix). The median percentages of weight loss from start of RT to completion of RT were 9.1% (IQR 6.7–11.2) in Arm 1 and 6.6% (IQR 2.6–8.1) in Arm 2.
Table 3.
Adverse events (AEs) that occurred during induction therapy
| Induction Arm 1: nab-Paclitaxel + cisplatin (n = 40) | Induction Arm 2: nab-Paclitaxel (n = 40) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
|||||||||
| Overall | Grade 1–2 | Grade 3 | Grade 4 | Grade 5 | Overall | Grade 1–2 | Grade 3 | Grade 4 | Grade 5 | |
| Hematological | ||||||||||
| Hemoglobin | 40 (100%) | 36 (90%) | 3 (8%) | 1 (3%) | 0 | 35 (88%) | 33 (83%) | 2 (5%) | 0 | 0 |
| Leukocytes (WBC) | 27 (68%) | 24 (60%) | 2 (5%) | 1 (3%) | 0 | 25 (63%) | 20 (50%) | 3 (8%) | 2 (5%) | 0 |
| Neutrophils (ANC) | 27 (68%) | 16 (40%) | 9 (23%) | 2 (5%) | 0 | 21 (53%) | 14 (35%) | 4 (10%) | 3 (8%) | 0 |
| Platelets | 16 (40%) | 16 (40%) | 0 | 0 | 0 | 3 (8%) | 3 (8%) | 0 | 0 | 0 |
| Non-hematologic | ||||||||||
| Fatigue | 24 (60%) | 24 (60%) | 0 | 0 | 0 | 21 (53%) | 20 (50%) | 1 (3%) | 0 | 0 |
| Dysphagia | 20 (50%) | 19 (48%) | 1 (3%) | 0 | 0 | 25 (63%) | 23 (58%) | 2 (5%) | 0 | 0 |
| Nausea | 18 (45%) | 18 (45%) | 0 | 0 | 0 | 14 (35%) | 14 (35%) | 0 | 0 | 0 |
| Voice changes | 16 (40%) | 16 (40%) | 0 | 0 | 0 | 22 (55%) | 22 (55%) | 0 | 0 | 0 |
| Hypertension | 0 | 0 | 0 | 0 | 0 | 16 (40%) | 15 (38%) | 1 (3%) | 0 | 0 |
| Alopecia | 15 (38%) | 15 (38%) | 0 | 0 | 0 | 14 (35%) | 14 (35%) | 0 | 0 | 0 |
| Constipation | 14 (35%) | 14 (35%) | 0 | 0 | 0 | 14 (35%) | 14 (35%) | 0 | 0 | 0 |
| Cough | 14 (35%) | 14 (35%) | 0 | 0 | 0 | 5 (13%) | 5 (13%) | 0 | 0 | 0 |
| Tumor pain | 14 (35%) | 14 (35%) | 0 | 0 | 0 | 10 (25%) | 7 (18%) | 3 (8%) | 0 | 0 |
| Diarrhea | 13 (33%) | 12 (30%) | 1 (3%) | 0 | 0 | 7 (18%) | 7 (18%) | 0 | 0 | 0 |
| Hyponatremia | 11 (28%) | 11 (28%) | 0 | 0 | 0 | 14 (35%) | 12 (30%) | 2 (5%) | 0 | 0 |
| Sinus tachycardia | 11 (28%) | 10 (25%) | 1 (3%) | 0 | 0 | 5 (13%) | 5 (13%) | 0 | 0 | 0 |
| Tinnitus | 11 (28%) | 11 (28%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Hypocalcemia | 10 (25%) | 9 (23%) | 0 | 1 (3%) | 0 | 6 (15%) | 6 (15%) | 0 | 0 | 0 |
| Other—pain | 10 (25%) | 9 (23%) | 1 (3%) | 0 | 0 | 4 (10%) | 2 (5%) | 1 (3%) | 1 (3%) | 0 |
| Creatinine | 9 (23%) | 7 (18%) | 1 (3%) | 1 (3%) | 0 | 14 (35%) | 14 (35%) | 0 | 0 | 0 |
| Dyspepsia | 9 (23%) | 9 (23%) | 0 | 0 | 0 | 5 (13%) | 5 (13%) | 0 | 0 | 0 |
| Hypoalbuminemia | 9 (23%) | 9 (23%) | 0 | 0 | 0 | 12 (30%) | 12 (30%) | 0 | 0 | 0 |
| Dyspnea | 0 | 0 | 0 | 0 | 0 | 10 (25%) | 10 (25%) | 0 | 0 | 0 |
| Anorexia | 8 (20%) | 8 (20%) | 0 | 0 | 0 | 7 (18%) | 6 (15%) | 1 (3%) | 0 | 0 |
| Edema | 8 (20%) | 8 (20%) | 0 | 0 | 0 | 9 (23%) | 9 (23%) | 0 | 0 | 0 |
| Hyperglycemia | 8 (20%) | 5 (13%) | 3 (8%) | 0 | 0 | 9 (23%) | 7 (18%) | 2 (5%) | 0 | 0 |
| Weight loss | 8 (20%) | 8 (20%) | 0 | 0 | 0 | 9 (23%) | 7 (18%) | 2 (5%) | 0 | 0 |
| ALT | 7 (18%) | 7 (18%) | 0 | 0 | 0 | 8 (20%) | 8 (20%) | 0 | 0 | 0 |
| Dizziness | 7 (18%) | 7 (18%) | 0 | 0 | 0 | 4 (10%) | 4 (10%) | 0 | 0 | 0 |
| Dysgeusia | 7 (18%) | 7 (18%) | 0 | 0 | 0 | 4 (10%) | 4 (10%) | 0 | 0 | 0 |
| Rash | 7 (18%) | 6 (15%) | 1 (3%) | 0 | 0 | 10 (25%) | 10 (25%) | 0 | 0 | 0 |
| Hypokalemia | 6 (15%) | 3 (8%) | 3 (8%) | 0 | 0 | 5 (13%) | 3 (8%) | 2 (5%) | 0 | 0 |
| Neuropathy—sensory | 6 (15%) | 6 (15%) | 0 | 0 | 0 | 11 (28%) | 10 (25%) | 1 (3%) | 0 | 0 |
| Dehydration | 5 (13%) | 2 (5%) | 3(8%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Hyperkalemia | 5 (13%) | 5 (13%) | 0 | 0 | 0 | 10 (25%) | 9 (23%) | 1 (3%) | 0 | 0 |
| Insomnia | 5 (13%) | 5 (13%) | 0 | 0 | 0 | 4 (10%) | 4 (10%) | 0 | 0 | 0 |
| Thrombosis/embolism | 5 (13%) | 4 (10%) | 1 (3%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Alkaline phosphatase | 4 (10%) | 4 (10%) | 0 | 0 | 0 | 6 (15%) | 6 (15%) | 0 | 0 | 0 |
| AST | 4 (10%) | 4 (10%) | 0 | 0 | 0 | 9 (23%) | 9 (23%) | 0 | 0 | 0 |
| Hemorrhage | 4 (10%) | 3 (8%) | 1 (3%) | 0 | 0 | 6 (15%) | 6 (15%) | 0 | 0 | 0 |
| Hypomagnesemia | 4 (10%) | 3 (8%) | 1 (3%) | 0 | 0 | 6 (15%) | 6 (15%) | 0 | 0 | 0 |
| Mucositis | 4 (10%) | 4 (10%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Vomiting | 0 | 0 | 0 | 0 | 0 | 5 (13%) | 5 (13%) | 0 | 0 | 0 |
| INR | 2 (5%) | 1 (3%) | 1 (3%) | 0 | 0 | 5 (13%) | 5 (13%) | 0 | 0 | 0 |
| Infection: skin | 1 (3%) | 0 | 1 (3%) | 0 | 0 | 4 (10%) | 3 (8%) | 1 (3%) | 0 | 0 |
| Hearing | 0 | 0 | 0 | 0 | 0 | 6 (15%) | 6 (15%) | 0 | 0 | 0 |
| Arthritis | 0 | 0 | 0 | 0 | 0 | 4 (10%) | 4 (10%) | 0 | 0 | 0 |
| Dry mouth | 0 | 0 | 0 | 0 | 0 | 4 (10%) | 4 (10%) | 0 | 0 | 0 |
| Proteinuria | 0 | 0 | 0 | 0 | 0 | 4 (10%) | 3 (8%) | 1 (3%) | 0 | 0 |
| Ulceration | 0 | 0 | 0 | 0 | 0 | 4 (10%) | 4 (10%) | 0 | 0 | 0 |
Discussion
Among patients with LA-HNSCC, achievement of a favorable (cCR or cPR) tumor response at the primary site, assessed by clinical examination, after induction cisplatin and 5-fluorouracil correlated with improved disease control after radiation therapy [25]. Patients who experienced cCR had the best outcomes; however, the cCR rate with induction cisplatin and 5-fluorouracil was only 20% [26]. We reported that the cCR rate at the primary site after induction APF was 77%, and the addition of cetuximab to APF did not improve this outcome [18-20]. In this trial, we evaluated the contribution of each agent in the APF regimen to tumor response. The cCR rate at the primary site after nab-paclitaxel and cisplatin was 70%, similar to APF; however, the cCR rate after nab-paclitaxel monotherapy was 20%, lower than with APF. The favorable clinical response rate at the primary tumor site was 98% after nab-paclitaxel and cisplatin and 90% after nab-paclitaxel monotherapy. These results suggest that both nab-paclitaxel and cisplatin significantly contributed to the achievement of tumor response at the primary site, whereas, 5-fluorouracil added little, if any benefit.
T stage classification and HPV status are key factors that affect tumor response to current induction regimens. The cCR rates after current induction chemotherapy are higher in smaller tumors (lower T classification) and in HPV-positive OPSCC [25]. In Arm 1 of our trial, the cCR rates after two cycles of nab-paclitaxel and cisplatin were high (54–86%) across all T classifications tested, and in both HPV-positive OPSCC (72%) and HPV-negative HNSCC (64%). In Arm 2, the cCR rates after two cycles of nab-paclitaxel were similar (11–26%) across all T classifications tested; however, the cCR rate was higher in HPV-positive OPSCC (35%) than in HPV-negative HNSCC (9%).
Others have reported the cCR rates at the primary site with paclitaxel and cisplatin- based induction regimens. Cross-trial comparisons have limitations but provide a frame of reference for our results. In our trial, the patient sample in Arm 1 was heterogeneous but enriched with larger tumors (68% T3 or T4 classification) and HPV-positive OPSCC (73%). A randomized trial with a heterogeneous patient sample reported the cCR rate after paclitaxel, cisplatin, and 5-fluorouracil was 33%, and after cisplatin and 5-fluorouracil was 14% [27]. The ECOG1308 trial reported a high cCR rate (70%) after paclitaxel, cisplatin, and cetuximab; however, most patients had small tumors (T1 or T2 classification) and all had HPV-positive OPSCC [28].
Because LRC after CRT inversely correlated with pre-treatment tumor volume [4], we hypothesized that reduction of tumor volume with nab-paclitaxel-based therapy given before CRT would increase LRC. Significant reduction of tumor volume, reflected by the surrogate of overall tumor response assessed by clinical examination, RECIST1.1 and PERCIST, occurred after two cycles of nab-paclitaxel-based therapy. In Arm 1, the LRC at two years was 89% in HPV-negative HNSCC and 100% in HPV-positive OPSCC. In Arm 2, the LRC at two years was 72% in HPV-negative HNSCC and 85% in HPV-positive OPSCC. Contemporary reports of CisRT showed LRC of 48% in HPV-negative HNSCC and 80–90% in HPV-positive OPSCC [29, 30]. Reports of CetuxRT showed LRC of 32% in HPV-negative OPSCC and 83% in HPV-positive OPSCC [29, 31]. Randomized trials are required to determine if the addition of nab-paclitaxel-based induction to CRT improves LRC over CRT alone.
The proportion of patients who experienced grade 3–4 AEs during induction therapy was 58% with nab-paclitaxel and cisplatin and 43% with nab-paclitaxel monotherapy. These results were not lower than with APF (40%) [19]. Patients in Arm 2 had significant co-morbidities that may have contributed to the frequency of some of the AEs observed in this arm. However, the proportion of patients who experienced grade 3–4 diarrhea and/or oral ulcers during induction therapy was 3% in each arm, whereas, it was 13% with APF. The higher frequency of gastrointestinal toxicities with APF may be due to 5-FU [19].
In patients with non-nasopharyngeal LA-HNSCC who have good organ function and performance status, the standard-of-care non-surgical treatment is CisRT. Although induction chemotherapy has a role as an organ preservation strategy, no clear survival benefit from current induction regimens emerged from randomized trials [5-8]. Paclitaxel was tested in randomized trials of induction therapy, but nab-paclitaxel has not. These two drugs are not equivalent [13-17]. In our trial, nab-paclitaxel and cisplatin followed by CisRT resulted in PFS and OS at 2 years of 90 and 100% in HPV-positive OPSCC and 82 and 89% in HPV-negative HNSCC. Contemporary reports of CisRT showed rates of PFS and OS of 78 and 85% in HPV-positive OPSCC and 23 and 30% in HPV-negative HNSCC [29, 30]. A comparison of Arm 1 of our trial to contemporary reports is limited by the small number of patients within each HPV strata.
Induction therapy had not been tested in patients who are unsuitable candidates for cisplatin, because current induction regimens would not be safe to administer. In this trial, nab-paclitaxel monotherapy was tolerable in most of these patients, and delivery of nab-paclitaxel and CetuxRT was favorable. nab-paclitaxel monotherapy followed by CetuxRT resulted in PFS and OS at 2 years of 53 and 71% in HPV-positive OPSCC and 54 and 67% in HPV-negative HNSCC. In the IMCL-9815 trial, CetuxRT resulted in PFS and OS of 82 and 88% in HPV-positive OPSCC and 29 and 42% in HPV-negative OPSCC [31]. A comparison of Arm 2 of our trial to IMCL-9815 is hampered by key differences in patient and tumor characteristics which would be expected to favor better survival outcomes for IMCL-9815.
There are limitations of this trial. The study design was a non-randomized, phase 2 trial without a direct comparison arm (APF, other induction regimen, or CRT alone). The sample size of each arm was insufficient to perform adequately powered subset analyses. The primary endpoint, cCR, is a semi-quantitative measure of tumor response assessed by clinical examination. Not all accept this assessment method as the “gold standard”. However, clinical tumor response has been widely used since the inception of induction therapy as an organ preservation strategy because the primary tumor site and its variable dimensions are easy to directly visualize and palpate on clinical examination and the broad response bundles are easily reproducible. In contrast, the measurement of primary tumor site on CT scan is often challenging due to poorly defined, irregular tumor borders and technical issues like variations in contrast enhancement with repeated CT scans. The duration of follow-up for Arm 2 was not mature. CetuxRT was used in Arm 2; other investigators may prefer a carboplatin-based regimen with RT in patients who are unsuitable candidates for cisplatin. One may argue that a comparison of the efficacy of the induction regimen used in Arm 2 of this trial to the historical reference (APF) is unfair because patients in Arm 2 were older and had poorer performance status and higher T stage classification. However, delivery of nab-paclitaxel was similar in both cohorts and the cCR rate at the primary site with both induction regimens showed no clear association with T classification.
Conclusion
In this multicenter, phase 2 trial, both nab-paclitaxel and cisplatin significantly contributed to achievement of a cCR at the primary tumor site. In comparing the results of this trial with the APF regimen, 5-fluorouracil appears to add little if any benefit to nab-paclitaxel and cisplatin [19]. Based on the favorable LRC and survival outcomes of each arm of the trial, further investigation of nab-paclitaxel as a therapeutic agent in LA-HNSCC is warranted. Combinations of nab-paclitaxel with immunotherapy are of particular interest given the priming effects of nab-paclitaxel on the immune system [32]. Studies are underway, which are evaluating the efficacy of this novel combination in HNSCC (NCT03174275; NCT03975270; NCT03107182).
Supplementary Material
Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s12032-021-01479-w.
Acknowledgements
The Siteman Cancer Center is supported in part by NCI Cancer Center Support (Grant number P30 CA91842). We acknowledge funding from Celgene for this project and the support of Tin Jin Ong MD, Kathy Amiri PhD, Jeff Eskew PhD, Monserrat Valles PhD, Jacqueline Jesse PharmD, and Daniel Read PhD. We thank the patients who participated, the head and neck research team at Washington University, Kim Trinkaus PhD, the Alvin Siteman Cancer Center (Washington University, St Louis, Missouri, USA) and Barnes-Jewish Hospital (St Louis) for use of the Tissue Procurement Center, and the Center for Biomedical Informatics.
Funding
This work was supported in part by Grant funding from Celgene. Clinical trial sponsorship by Celgene and Siteman Cancer Center in part by NCI Cancer Center Support Award Number P30 CA91842.
Footnotes
Conflict of interest JL, MD, JR, RP, PP, JL, HG, PN, and MG have no disclosures to report. PO reports personal fees from Bristol Myers, Merck, and Eisai, outside the submitted work. RJ reports personal fees from Intuitive surgical, outside the submitted work. KP reports personal fees from Merck, Genetech, and Boehringer Ingelheim, outside the submitted work. RS reports consulting support from Bristol Myers and institutional research support from Merck, Bristol Myers, Pfizer, Vyriad, Incyte, Genentech, Actuate, and Novartis, outside the submitted work. WS reports consulting support from Bristol Myers and Regeneron, outside the submitted work. JZ reports personal fees from Summit Biolabs, Inc, outside the submitted work. WT reports other from Elekta, Inc (wife is employee. Radiotherapy hardware and software company. However, the department does not use those products). DA reports research funding from Celgene for the submitted work and consulting or scientific advisory board support from Pfizer, Eli Lilly, Merck, Celgene, Cue Biopharma, and institutional research support from Pfizer, Eli Lilly, Merck, Celgene/BMS, Novartis, AstraZeneca, Atara Bio, Blueprint Medicine, Celldex, Enzychem, Kura, Exelixis, Innate, Sensei, and Matrix Biomed, outside the submitted work.
Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent Informed consent was obtained from all individual participants included in the study. The study was reviewed and approved by the Washington University School of Medicine Internal Review Board, St. Louis, Missouri. Patients signed informed consent regarding publishing their data.
Presented in abstract form at the European Society of Medical Oncology Annual Meeting, 2018 and the Multidisciplinary Head and Neck Cancers Symposium, 2020
Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
- 1.Haddad RI, Shin DM. Recent advances in head and neck cancer. N Engl J Med. 2008;359:1143–54. [DOI] [PubMed] [Google Scholar]
- 2.Ahn MJ, D’Cruz A, Vermoken JB, et al. Clinical recommendations for defining platinum unsuitable head and neck cancer patient populations on chemoradiotherapy: a literature review. Oral Oncol. 2016;53:10–6. [DOI] [PubMed] [Google Scholar]
- 3.Baxi SS, O’Neill C, Sherman EJ, et al. Trends in chemoradiation use in elderly patients with head and neck cancer: changing treatment patterns with cetuximab. Head Neck. 2016;38:E165–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pak K, Cheon GJ, Nam HY, et al. Prognostic value of metabolic tumor volume and total lesion glycolysis in head and neck cancer: a systematic review and meta-analysis. J Nucl Med. 2014;55:884–90. [DOI] [PubMed] [Google Scholar]
- 5.Ghi MG, Paccagnella A, Ferrari D, et al. Induction TPF followed by concomitant treatment versus concomitant treatment alone in locally advanced head and neck cancer: a phase II-III trial. Ann Oncol. 2017;28(9):2206–12. [DOI] [PubMed] [Google Scholar]
- 6.Cohen EW, Karrison TG, Kocherginsky M, et al. Phase III randomized trial of induction chemotherapy in patients with N2 or N3 locally advanced head and neck cancer. J Clin Oncol. 2014;32(25):2735–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Geoffrois L, Martin L, De Raucourt D, et al. Induction chemotherapy followed by cetuximab radiotherapy is not superior to concurrent chemoradiotherapy for head and neck carcinomas: results of the GORTEC 2007–02 phase III randomized trial. J Clin Oncol. 2018;36(31):3077–83. [DOI] [PubMed] [Google Scholar]
- 8.Haddad R, O’Neill A, Rabinowits G, et al. Induction chemotherapy followed by concurrent chemoradiotherapy (sequential chemoradiotherapy) versus concurrent chemoradiotherapy alone in locally advanced head and neck cancer (PARADIGM): a randomised phase 3 trial. Lancet Oncol. 2013;14(3):257–64. [DOI] [PubMed] [Google Scholar]
- 9.Recouvreux MV, Commisso C. Macropinocytosis: a metabolic adaptation to nutrient stress in cancer. Front Endocrinol. 2017;8:261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Commisso C, Davidson SM, Soydaner-Azeloglu RD, et al. Macropinocytosis of protein is an amino acid supply route in Ras-transformed cells. Nature. 2013;497:633–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Nakase I, Kobayashi NB, Takatani-Nakase T, Yoshida T. Active macropinocytosis induction by stimulation of epidermal growth factor receptor and oncogenic Ras expression potentiates cellular uptake efficacy of exosomes. Sci Rep. 2015;5:10300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.The Cancer Genome Atlas Network. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature. 2015;517:576–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Desai N, Trieu V, Yao Z, et al. Increased antitumor activity, intratumor paclitaxel concentrations, and endothelial cell transport of cremophor-free, albumin-bound paclitaxel, ABI-007, compared with cremophor-based paclitaxel. Clin Cancer Res. 2006;12(4):1317–24. [DOI] [PubMed] [Google Scholar]
- 14.Gradishar WJ, Tjulandin S, Davidson N, et al. Phase III trial of nanoparticle albumin-bound paclitaxel compared with polyethylated castor oil-based paclitaxel in women with breast cancer. J Clin Oncol. 2005;23(31):7794–803. [DOI] [PubMed] [Google Scholar]
- 15.Socinski MA, Bondarenko I, Karaseva NA, et al. Weekly nab-paclitaxel in combination with carboplatin versus solvent-based paclitaxel plus carboplatin as first-line therapy in patients with advanced non-small-cell lung cancer: final results of a phase III trial. J Clin Oncol. 2012;30(17):2055–62. [DOI] [PubMed] [Google Scholar]
- 16.Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18):1691–703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hersh EM, Del Vecchio M, Brown MP, et al. A randomized, controlled phase III trial of nab-paclitaxel versus dacarbazine in chemotherapy-naïve patients with metastatic melanoma. Ann Oncol. 2015;26(11):2267–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Adkins D, Ley J, Trinkaus K, et al. A phase 2 trial of induction nab-paclitaxel and cetuximab given with cisplatin and 5-fluorouracil followed by concurrent cisplatin and radiation for locally advanced squamous cell carcinoma of the head and neck. Cancer. 2013;119(4):766–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Adkins D, Ley J, Michel L, et al. nab-Paclitaxel, cisplatin, and 5-fluorouracil followed by concurrent cisplatin and radiation for head and neck squamous cell carcinoma. Oral Oncol. 2016;61:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Adkins D, Ley J, Oppelt P, et al. nab-Paclitaxel-based induction chemotherapy with or without cetuximab for locally advanced head and neck squamous cell carcinoma. Oral Oncol. 2017;72:26–31. [DOI] [PubMed] [Google Scholar]
- 21.Weiss J, Gilbert J, Deal AM, et al. Induction chemotherapy with carboplatin, nab-paclitaxel and cetuximab for at least N2b nodal status or surgically unresectable squamous cell carcinoma of the head and neck. Oral Oncol. 2018;84:46–51. [DOI] [PubMed] [Google Scholar]
- 22.Seiwert TY, Foster CC, Blair EA, et al. OPTIMA: a phase II dose and volume de-escalation trial for human papillomavirus-positive oropharyngeal cancer. Ann Oncol. 2019;30(2):297–302. [DOI] [PubMed] [Google Scholar]
- 23.Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000;92:205–16. [DOI] [PubMed] [Google Scholar]
- 24.Wahl RL, Jacene H, Kasamon Y, Lodge MA. From RECIST to PERCIST: evolving considerations for PET response criteria in solid tumors. J Nucl Med. 2009;50(Suppl. 1):122S–S150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ensley JF, Jacobs JR, Weaver A, et al. Correlation between response to cisplatinum-combination chemotherapy and subsequent radiotherapy in previously untreated patients with advanced squamous cell cancers of the head and neck. Cancer. 1984;54(5):811–4. [DOI] [PubMed] [Google Scholar]
- 26.Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med. 2003;349(22):2091–8. [DOI] [PubMed] [Google Scholar]
- 27.Hitt R, Lopez-Pousa A, Martinez-Trufero J, et al. Phase III study comparing cisplatin plus fluorouracil to paclitaxel, cisplatin, and fluorouracil induction chemotherapy followed by chemoradiotherapy in locally advanced head and neck cancer. J Clin Oncol. 2005;23(34):8636–45. [DOI] [PubMed] [Google Scholar]
- 28.Marur S, Li S, Cmelak AJ, et al. E1308: phase II trial of induction chemotherapy followed by reduced-dose radiation and weekly cetuximab in patients with HPV-associated resectable squamous cell carcinoma of the oropharynx—ECOG-ACRIN Cancer Research Group. J Clin Oncol. 2017;35(5):490–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Gillison ML, Trotti AM, Harris J, et al. Radiotherapy plus cetuximab or cisplatin in human papillomavirus-positive oropharyngeal cancer (NRG Oncology RTOG 1016): a randomised, multicentre, non-inferiority trial. Lancet. 2019;393(10166):40–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Nguyen-Tan PF, Zhang Q, Ang KK, et al. Randomized phase III trial to test accelerated versus standard fractionation in combination with concurrent cisplatin for head and neck carcinomas in the Radiation Therapy Oncology Group 0129 trial: Long-term report of efficacy and toxicity. J Clin Oncol. 2014;32(34):3858–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rosenthal DI, Harari PM, Giralt J, et al. Association of human papillomavirus and p16 status with outcomes in the IMCL-9815 phase III registration trial for patients with locoregionally advanced oropharyngeal squamous cell carcinoma of the head and neck treated with radiotherapy with or without cetuximab. J Clin Oncol. 2016;34(12):1300–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Soliman HH. nab-Paclitaxel as a potential partner with checkpoint inhibitors in solid tumors. Onco Targets Ther. 2017;10:101–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.