Abstract
Neoadjuvant bevacizumab with modified FOLFOX7 without radiation failed to meet the goal of pathological complete response rate; however, the low number of recurrence and disease‐free survival in this population, with predominantly stage III, is encouraging and worth further exploration.
The racial distribution of the patient population, as well as a wait time of more than 4 weeks after last chemotherapy, may have contributed to the findings.
Combination chemotherapy in lieu of radiation in rectal adenocarcinoma is under exploration in multiple trials. We evaluated the efficacy of neoadjuvant FOLFOX + bevacizumab in patients (pts) with clinical stage II and III disease.
Pts received six cycles of bevacizumab (5 mg/kg) and modified FOLFOX7 (oxaliplatin 85 mg/m2, leucovorin 20 mg/m2, and fluorouracil [5‐FU] 2,400 mg/m2). Surgical resection was performed 6–8 weeks after completion of treatment and upon confirmation of nonmetastatic disease. We employed a Simon two‐stage design and required three pathological complete responses (pCR) in the first 18 pts, with a prespecified pCR rate of 25% before moving to the next stage.
Seventeen pts enrolled; 65% at stage III. Median age was 57 (35–79), 65% were male, 47% were Hispanic, 35% were white, and 18% were Asian. All pts but one completed six cycles of therapy. One pCR was observed (6%), and 11 of 17 (65%) pts had pathological downstaging. One patient experienced systemic recurrence and remains on treatment. Probability of disease‐free survival (DFS) at 5 years is 0.94 (SE, 0.06).
The study failed to meet the required three pCRs in the first 18 pts. The DFS in this population is encouraging and supports the hypothesis that select pts with rectal cancer may be spared from radiation.
Discussion
In the U.S., the current standard of treatment for resectable stage II/III rectal cancer is surgical resection combined with neoadjuvant chemoradiation (chemoRT) and adjuvant chemotherapy. Although multiple studies have shown that neoadjuvant chemoRT reduces local recurrence, it is not associated with improvement in overall survival (OS). Additionally, chemoRT is associated with short‐ and long‐term toxicities such as pain, proctitis, cystitis, bowel obstruction, fistulas, fertility issues, sexual dysfunction, and second malignancies. Thus, efforts to identify newer strategies with reduced toxicities is of paramount importance.
In a small study, neoadjuvant FOLFOX appeared to be safe and resulted in pCR in a subset of patients [1]. A number of preclinical and clinical studies supported a role for bevacizumab in neoadjuvant setting. Two studies using FOLFOX and XELOX with bevacizumab reported a pCR rate of 20%–25% [2, 3].
This phase II study was designed to explore the efficacy of neoadjuvant bevacizumab with mFOLFOX7 (BEVmFOLFOX7) in stage II/III rectal cancer in a diverse patient population. Patients with histologically confirmed rectal adenocarcinoma underwent staging with CT scan of chest, abdomen, and pelvis; endorectal ultrasound (ERUS), and/or magnetic resonance imaging (MRI) of pelvis. Patients were required to have acceptable organ function and ECOG performance status of 0–1. All patients were planned to receive six cycles of BEVmFOLFOX7, followed by restaging with CT scan and ERUS to confirm no progression and subsequently underwent surgical resection 6–8 weeks after completion of their treatment.
Between September of 2013 through April of 2017, 17 patients were enrolled. Sixteen patients received all six cycles of treatment without unexpected toxicities; only one patient developed neutropenic fever after cycle 5. All patients underwent R0 resection with no wound dehiscence, prolonged hospitalization, or reoperation. Patients received adjuvant chemotherapy and chemoRT at the discretion of their treating physician.
Our study failed to meet the prespecified pCR rate of 25% and stopped further accrual. All patients had negative margins on surgical specimen (R0) and significant tumor downstaging, with stage III representing 29% of postresection population. With a median follow‐up of 44.8 months (27.4–63.9 months), one systemic recurrence was observed at 11 months from treatment initiation (Fig. 1).
Overall and disease‐free survival. Overall (red) and disease‐free (blue) survival with median follow‐up of 44.8 (27.4–63.9) months.
Although it is difficult to pinpoint the reasons for a significantly lower rates of pCR than expected, delayed time to surgery after the last chemotherapy cycle is the potential culprit of the observed outcome. Given the transient effects of chemotherapy on cancer cells, long delays could result in repopulation of cells, as suggested in the neoadjuvant treatment of breast cancer [4]. Additionally, given that bevacizumab interferes with the tumor microenvironment, race and ethnicity may dictate different levels of efficacy [5]. Our study enrolled a higher percentage of Hispanic patients (47%) than reported for most trials. Lack of enrollment of minorities on clinical trials limits our ability to assess disparities based on race in the effectiveness of biological therapies such as bevacizumab.
Trial Information
| Disease | Rectal cancer |
| Stage of Disease/Treatment | Neoadjuvant |
| Prior Therapy | None |
| Type of study | Phase II, single arm |
| Primary Endpoint | Rate of pCR after neoadjuvant therapy in pathological specimen |
| Secondary Endpoints | Tumor regression, rate of locoregional recurrence, incidence and nature of adverse events, progression‐free survival |
| Additional Details of Endpoints or Study Design | |
| Study Design: This was a single‐institution, single‐arm phase II trial conducted at the University of Southern California‐Norris Comprehensive Cancer Center. All patients received six cycles of BEVmFOLFOX, where each cycle consisted of bevacizumab 5 mg/kg, oxaliplatin 85 mg/m2, leucovorin 20 mg/m2, and 5‐FU 2,400 mg/m2 over 46 hours and was given every 2 weeks. Patients had follow‐up staging (including ERUS) 2–4 weeks after completion of cycle six to exclude metastatic and/or progressive disease. This was a Simon two‐stage design. Upon meeting the required number of pCR in the first stage (3/18), the study would have enrolled another 25 patients. We enrolled 17 patients with only one pCR and thus closed the study. | |
| Patients: The eligibility criteria included patients with histologically confirmed adenocarcinoma of the rectum located 5–15 cm from anal verge; age ≥ 18 years with life expectancy more than 12 weeks; no prior therapy and a candidate for surgical resection; tumor lesion staging cT2 N+ or cT3–T4 N0 or N+ as assessed by endorectal US or MRI; no evidence of distant metastatic disease; and Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. Adequate hematological, hepatic, and renal function was required, described as absolute neutrophil count ≥1,500 per mm3 and platelet count ≥100,000 per mm3 and hemoglobin ≥9 g/dL (may be transfused to maintain or exceed this level), total bilirubin <1.5 × upper limit of normal (ULN), and aspartate aminotransferase and alanine aminotransferase <2.5 × ULN, with calculated creatinine clearance according to the formula of Cockroft and Gault ≥50 mL/min urine proteinuria <2+ and international normalized ratio ≤1.5 and activated prothrombin time ≤1.5 × ULN within 7 days prior to enrollment. | |
| Investigator's Analysis | Level of activity did not meet planned endpoint |
| Disease | Rectal cancer |
| Stage of Disease/Treatment | Neoadjuvant |
| Prior Therapy | None |
| Type of study | Phase II, single arm |
| Primary Endpoint | Rate of pCR after neoadjuvant therapy in pathological specimen |
| Secondary Endpoints | Tumor regression, rate of locoregional recurrence, incidence and nature of adverse events, progression‐free survival |
| Additional Details of Endpoints or Study Design | |
| Study Design: This was a single‐institution, single‐arm phase II trial conducted at the University of Southern California‐Norris Comprehensive Cancer Center. All patients received six cycles of BEVmFOLFOX, where each cycle consisted of bevacizumab 5 mg/kg, oxaliplatin 85 mg/m2, leucovorin 20 mg/m2, and 5‐FU 2,400 mg/m2 over 46 hours and was given every 2 weeks. Patients had follow‐up staging (including ERUS) 2–4 weeks after completion of cycle six to exclude metastatic and/or progressive disease. This was a Simon two‐stage design. Upon meeting the required number of pCR in the first stage (3/18), the study would have enrolled another 25 patients. We enrolled 17 patients with only one pCR and thus closed the study. | |
| Patients: The eligibility criteria included patients with histologically confirmed adenocarcinoma of the rectum located 5–15 cm from anal verge; age ≥ 18 years with life expectancy more than 12 weeks; no prior therapy and a candidate for surgical resection; tumor lesion staging cT2 N+ or cT3–T4 N0 or N+ as assessed by endorectal US or MRI; no evidence of distant metastatic disease; and Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. Adequate hematological, hepatic, and renal function was required, described as absolute neutrophil count ≥1,500 per mm3 and platelet count ≥100,000 per mm3 and hemoglobin ≥9 g/dL (may be transfused to maintain or exceed this level), total bilirubin <1.5 × upper limit of normal (ULN), and aspartate aminotransferase and alanine aminotransferase <2.5 × ULN, with calculated creatinine clearance according to the formula of Cockroft and Gault ≥50 mL/min urine proteinuria <2+ and international normalized ratio ≤1.5 and activated prothrombin time ≤1.5 × ULN within 7 days prior to enrollment. | |
| Investigator's Analysis | Level of activity did not meet planned endpoint |
Drug Information
| Drug 1 | |
| Generic/Working Name | Bevacizumab |
| Trade Name | Avastin |
| Company Name | Genentech |
| Drug Type | Antibody |
| Dose | 5 mg/kg |
| Route | IV |
| Schedule of Administration | |
| Every 2 weeks. Study schema: six cycles of BEVmFOFLOX and had follow‐up staging (including ERUS) 2–4 weeks after completion of cycle six to exclude metastatic and/or progressive disease. Each cycle of therapy included bevacizumab 5 mg/kg, oxaliplatin 85 mg/m2, leucovorin 20 mg/m2, and 5‐FU 2,400 mg/m2 over 46 hours and was given every 2 weeks. | |
| Drug 2 | |
| Generic/Working Name | Folinic acid (Leucovorin) |
| Dose | 20 mg/m2 |
| Route | IV |
| Drug 3 | |
| Generic/Working Name | 5‐FU |
| Drug Type | Small molecule |
| Drug Class | Antimetabolite |
| Dose | 2,400 mg/m2 |
| Route | IV |
| Schedule of Administration | Infused over 46 hours |
| Drug 4 | |
| Generic/Working Name | Oxaliplatin |
| Drug Type | Small molecule |
| Drug Class | Platinum compound |
| Dose | 85 mg/m2 |
| Route | IV |
| Drug 1 | |
| Generic/Working Name | Bevacizumab |
| Trade Name | Avastin |
| Company Name | Genentech |
| Drug Type | Antibody |
| Dose | 5 mg/kg |
| Route | IV |
| Schedule of Administration | |
| Every 2 weeks. Study schema: six cycles of BEVmFOFLOX and had follow‐up staging (including ERUS) 2–4 weeks after completion of cycle six to exclude metastatic and/or progressive disease. Each cycle of therapy included bevacizumab 5 mg/kg, oxaliplatin 85 mg/m2, leucovorin 20 mg/m2, and 5‐FU 2,400 mg/m2 over 46 hours and was given every 2 weeks. | |
| Drug 2 | |
| Generic/Working Name | Folinic acid (Leucovorin) |
| Dose | 20 mg/m2 |
| Route | IV |
| Drug 3 | |
| Generic/Working Name | 5‐FU |
| Drug Type | Small molecule |
| Drug Class | Antimetabolite |
| Dose | 2,400 mg/m2 |
| Route | IV |
| Schedule of Administration | Infused over 46 hours |
| Drug 4 | |
| Generic/Working Name | Oxaliplatin |
| Drug Type | Small molecule |
| Drug Class | Platinum compound |
| Dose | 85 mg/m2 |
| Route | IV |
Patient Characteristics
| Number of Patients, Male | 11 |
| Number of Patients, Female | 6 |
| Stage | Stage II: 6 (35%); stage III: 11 (65%) |
| Age | Median (range): 57 (35–79) |
| Number of Prior Systemic Therapies | Median: 0 |
| Performance Status: ECOG | 0 — 0 1 — 17 2 — 0 3 — 0 Unknown — |
| Other | White: 6 (35%); Hispanic: 8 (47%); Asian: 3 (18%) |
| Cancer Types or Histologic Subtypes | Adenocarcinoma, 17 |
| Number of Patients, Male | 11 |
| Number of Patients, Female | 6 |
| Stage | Stage II: 6 (35%); stage III: 11 (65%) |
| Age | Median (range): 57 (35–79) |
| Number of Prior Systemic Therapies | Median: 0 |
| Performance Status: ECOG | 0 — 0 1 — 17 2 — 0 3 — 0 Unknown — |
| Other | White: 6 (35%); Hispanic: 8 (47%); Asian: 3 (18%) |
| Cancer Types or Histologic Subtypes | Adenocarcinoma, 17 |
Primary Assessment Method
| Title | Pathological response to neoadjuvant therapy |
| Number of Patients Enrolled | 17 |
| Number of Patients Evaluable for Toxicity | 17 |
| Number of Patients Evaluated for Efficacy | 17 |
| Evaluation Method | Pathologic complete response (pCR) |
| Response Assessment CR | n = 1 (6%) |
| Response Assessment OTHER | n = 16 (94%) |
| Outcome Notes | |
| By clinical staging, 65% of the study population had stage III at the time of initiation of neoadjuvant therapy. By pathological staging and after completion of neoadjuvant therapy, only 29% of the study population had stage III. | |
| Title | Pathological response to neoadjuvant therapy |
| Number of Patients Enrolled | 17 |
| Number of Patients Evaluable for Toxicity | 17 |
| Number of Patients Evaluated for Efficacy | 17 |
| Evaluation Method | Pathologic complete response (pCR) |
| Response Assessment CR | n = 1 (6%) |
| Response Assessment OTHER | n = 16 (94%) |
| Outcome Notes | |
| By clinical staging, 65% of the study population had stage III at the time of initiation of neoadjuvant therapy. By pathological staging and after completion of neoadjuvant therapy, only 29% of the study population had stage III. | |
Secondary Assessment Method
| Title | Rate of Locoregional Recurrence |
| Number of Patients Enrolled | 17 |
| Number of Patients Evaluable for Toxicity | 17 |
| Number of Patients Evaluated for Efficacy | 17 |
| Response Assessment OTHER | n = 0 (0%) |
| Outcome Notes | There was one systemic recurrence (6%) but no locoregional recurrence. |
| Title | Tumor downstaging |
| Number of Patients Enrolled | 17 |
| Evaluation Method | Tumor stage after complete neoadjuvant therapy by ERUS |
| Outcome Notes | Stage 0: 1 (7%) Stage I: 2 (13%) Stage II: 7 (47%) Stage III: 5 (33%) Not Done: 2 |
| Title | Rate of Locoregional Recurrence |
| Number of Patients Enrolled | 17 |
| Number of Patients Evaluable for Toxicity | 17 |
| Number of Patients Evaluated for Efficacy | 17 |
| Response Assessment OTHER | n = 0 (0%) |
| Outcome Notes | There was one systemic recurrence (6%) but no locoregional recurrence. |
| Title | Tumor downstaging |
| Number of Patients Enrolled | 17 |
| Evaluation Method | Tumor stage after complete neoadjuvant therapy by ERUS |
| Outcome Notes | Stage 0: 1 (7%) Stage I: 2 (13%) Stage II: 7 (47%) Stage III: 5 (33%) Not Done: 2 |
Adverse Events
| All Cycles | |||||||
|---|---|---|---|---|---|---|---|
| Name | NC/NA | 1 | 2 | 3 | 4 | 5 | All Grades |
| Anemia | 70% | 18% | 12% | 0% | 0% | 0% | 30% |
| Febrile neutropenia | 94% | 0% | 0% | 6% | 0% | 0% | 6% |
| Constipation | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Diarrhea | 59% | 29% | 12% | 0% | 0% | 0% | 41% |
| Dry mouth | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Dyspepsia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Mucositis oral | 76% | 24% | 0% | 0% | 0% | 0% | 24% |
| Nausea | 6% | 94% | 0% | 0% | 0% | 0% | 94% |
| Oral hemorrhage | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Rectal hemorrhage | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Vomiting | 76% | 18% | 6% | 0% | 0% | 0% | 24% |
| Fatigue | 35% | 59% | 6% | 0% | 0% | 0% | 65% |
| Neutrophil count decreased | 70% | 6% | 12% | 0% | 12% | 0% | 30% |
| Platelet count decreased | 59% | 35% | 6% | 0% | 0% | 0% | 41% |
| White blood cell decreased | 82% | 12% | 0% | 0% | 6% | 0% | 18% |
| Alanine aminotransferase increased | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Aspartate aminotransferase increased | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Salivary gland infection | 100% | 0% | 0% | 0% | 0% | 0% | 0% |
| Anorexia | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Hyponatremia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Bone pain | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Myalgia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Dizziness | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Dysgeusia | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Headache | 88% | 6% | 6% | 0% | 0% | 0% | 12% |
| Nervous system disorders, cold sensitivity | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Peripheral sensory neuropathy | 17% | 71% | 0% | 12% | 0% | 0% | 83% |
| Insomnia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Proteinuria | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Cough | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Epistaxis | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Voice alteration | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Skin and subcutaneous tissue disorders ‐ nail discoloration | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Palmar‐plantar erythrodysesthesia syndrome | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Hypertension | 64% | 12% | 24% | 0% | 0% | 0% | 36% |
| All Cycles | |||||||
|---|---|---|---|---|---|---|---|
| Name | NC/NA | 1 | 2 | 3 | 4 | 5 | All Grades |
| Anemia | 70% | 18% | 12% | 0% | 0% | 0% | 30% |
| Febrile neutropenia | 94% | 0% | 0% | 6% | 0% | 0% | 6% |
| Constipation | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Diarrhea | 59% | 29% | 12% | 0% | 0% | 0% | 41% |
| Dry mouth | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Dyspepsia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Mucositis oral | 76% | 24% | 0% | 0% | 0% | 0% | 24% |
| Nausea | 6% | 94% | 0% | 0% | 0% | 0% | 94% |
| Oral hemorrhage | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Rectal hemorrhage | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Vomiting | 76% | 18% | 6% | 0% | 0% | 0% | 24% |
| Fatigue | 35% | 59% | 6% | 0% | 0% | 0% | 65% |
| Neutrophil count decreased | 70% | 6% | 12% | 0% | 12% | 0% | 30% |
| Platelet count decreased | 59% | 35% | 6% | 0% | 0% | 0% | 41% |
| White blood cell decreased | 82% | 12% | 0% | 0% | 6% | 0% | 18% |
| Alanine aminotransferase increased | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Aspartate aminotransferase increased | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Salivary gland infection | 100% | 0% | 0% | 0% | 0% | 0% | 0% |
| Anorexia | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Hyponatremia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Bone pain | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Myalgia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Dizziness | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Dysgeusia | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Headache | 88% | 6% | 6% | 0% | 0% | 0% | 12% |
| Nervous system disorders, cold sensitivity | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Peripheral sensory neuropathy | 17% | 71% | 0% | 12% | 0% | 0% | 83% |
| Insomnia | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Proteinuria | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Cough | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Epistaxis | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Voice alteration | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Skin and subcutaneous tissue disorders ‐ nail discoloration | 94% | 6% | 0% | 0% | 0% | 0% | 6% |
| Palmar‐plantar erythrodysesthesia syndrome | 88% | 12% | 0% | 0% | 0% | 0% | 12% |
| Hypertension | 64% | 12% | 24% | 0% | 0% | 0% | 36% |
Abbreviation: NC/NA, no change from baseline/no adverse event
Serious Adverse Events
| Name | Grade | Attribution |
|---|---|---|
| Neutrophil count decreased | 4 | Probable |
| White blood cell decreased | 4 | Probable |
| Name | Grade | Attribution |
|---|---|---|
| Neutrophil count decreased | 4 | Probable |
| White blood cell decreased | 4 | Probable |
Assessment, Analysis, and Discussion
| Completion | Study completed |
| Investigator's Assessment | Level of activity did not meet planned endpoint |
| Completion | Study completed |
| Investigator's Assessment | Level of activity did not meet planned endpoint |
A total of 44,180 new cases of rectal cancer are expected in 2019, constituting 2.5% of all cancer diagnosis [6]. The incidence of rectal cancer in younger patients is on the rise, and thus efforts to reduce the toxicities of the treatment while maintaining effectiveness remains crucial in advancing the care of this population.
Surgery remains the mainstay of cure in patients with rectal cancer with resectable disease, whereas radiation and chemotherapy reduce the chances of local and systemic recurrence [7, 8]. Neoadjuvant chemoradiation followed by surgery and subsequent adjuvant chemotherapy has been established as the standard of care in the U.S. [9, 10]. Although chemoradiation has been beneficial in reducing local recurrence, it does not confer overall survival (OS) benefit, and risk of systemic recurrence in most patients with rectal cancer remains higher than local recurrence [11–13]. Furthermore, chemoradiation to the rectum comes with a myriad of short‐ and long‐term toxicities [14, 15]. Given the improved efficacy of combination chemotherapy and targeted agents, the added benefit of radiation is questioned.
The GEMCAD group reported a pathological complete response (pCR) rate of 20% in patients with T3 rectal cancer receiving XELOX and bevacizumab. Yet, the high rate of anastomotic leak was prohibitive of further development of this treatment [3]. Similarly, Schrag et al. reported that FOLFOX and bevacizumab without radiotherapy results in a pCR rate of 25% and 4‐year recurrence rate of 0% [2]. This study did not report additional anastomotic leak due to longer interval from bevacizumab to resection.
Our study was designed to evaluate FOLFOX and bevacizumab for 6 cycles (as opposed to 4 cycles in the Schrag study) in a diverse patient population and with more advanced disease than those included in the GEMCAD and Schrag studies (Table 1). Out of 17 patients recruited for our study, only 1 patient had pCR (6%) which is closer to the pCR reported by Koizumi et al. [1] (Table 2). Because of the single‐arm nature of this study, it is difficult to pinpoint the reasons for such a low pCR. However, it is worth mentioning that our study had a unique mix of racial backgrounds consisting of 47% Hispanic, 35% white, and 18% Asian. Bevacizumab, an antiangiogenic agent, interferes with host response to tumor. Benefit of bevacizumab can vary based on race and ethnicity [5, 16, 17]. The observed difference between our study and the GEMCAD and Schrage studies may be due to lack of benefit from bevacizumab in minorities. Although the Schrag study from Memorial Sloan Kettering does not list the study population's race, most Memorial Sloan Kettering patients are white (81%) [18].
Another possible reason for the low pCR could be the delay in surgical resection after completion of the chemotherapy. In the Memorial Sloan Kettering study, FOLFOX with bevacizumab was used for the first 4 cycles, followed by an additional 2 cycles of FOLFOX, allowing for resection promptly after completing the last rounds of chemotherapy. In our study, patients had 6–8 weeks when they were not undergoing any treatment; median time to surgery was 52 days (range, 35–82), equivalent to 7.43 weeks. This may mean that the longer the patient is off of FOLFOX, the more likely the cancer will have time to regrow, leading to decrease in observed pCR. In patients with breast cancer, a longer wait time from chemotherapy to surgery was shown to result in a lower pCR rate [4]. For future studies of this kind, the interval between chemotherapy and surgery should be kept to a minimum.
Endpoint selection remains critical to measuring the effect of an intervention in the trial population. pCR is defined as having no evidence of viable tumor on pathologic analysis; achieving pCR after nCRT is associated with improved RFS, OS, and sphincter preservation [19–22]. Yet, pCR may not be the most suitable endpoint for assessing the benefit of neoadjuvant chemotherapy in rectal cancer. We evaluated the neoadjuvant rectal (NAR) score in our population as a measure of effectiveness of treatment. [23] The median NAR score in our population was 15; 1 patient had a low NAR (6%), 4 had high NAR (24%), and the rest had intermediate NAR (Table 2). The DFS and OS in our group for the calculated NAR compares favorably with chemoradiation studies and supports the activity of the regimen in achieving cure and long‐term survival [24, 25].
This study was negative, and FOLFOX with bevacizumab should not be used outside of clinical trials. However, the findings of this study can help with the design of future studies both in terms of selection of endpoint and in prescribing the interval from chemotherapy to surgery.
Acknowledgments
This work was supported by Genentech and supported in part by NCI (grant number P30CA014089).
Disclosures
Afsaneh Barzi: Genentech, Merck, Bayer (RF); Heinz‐Josef Lenz: Bayer, Genentech, Merck (C/A, RF). The other authors indicated no financial relationships.
(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board
Tables
Demographic and baseline clinical characteristics
| Characteristics | n (%) |
|---|---|
| Total eligible patients enrolled | 17 (100) |
| Enrollment time | September 11, 2013–April 10, 2017 |
| Age at study entry, median (range), yr | 57 (35–79) |
| Duration from diagnosis to study entry, mo | |
| Median (range) | 2.0 (0.6–5.1) |
| Missing | 1 |
| Gender | |
| Female | 6 (35) |
| Male | 11 (65) |
| Race or ethnicity | |
| White | 6 (35) |
| Hispanic | 8 (47) |
| Asian | 3 (18) |
| Tumor clinical stage at study entry by ERUS | |
| II | 6 (35) |
| III | 11 (65) |
| Distance of tumor to anal verge, median (range), cm | 8 (5–14) |
| Characteristics | n (%) |
|---|---|
| Total eligible patients enrolled | 17 (100) |
| Enrollment time | September 11, 2013–April 10, 2017 |
| Age at study entry, median (range), yr | 57 (35–79) |
| Duration from diagnosis to study entry, mo | |
| Median (range) | 2.0 (0.6–5.1) |
| Missing | 1 |
| Gender | |
| Female | 6 (35) |
| Male | 11 (65) |
| Race or ethnicity | |
| White | 6 (35) |
| Hispanic | 8 (47) |
| Asian | 3 (18) |
| Tumor clinical stage at study entry by ERUS | |
| II | 6 (35) |
| III | 11 (65) |
| Distance of tumor to anal verge, median (range), cm | 8 (5–14) |
Abbreviation: ERUS, endorectal ultrasound.
Demographic and baseline clinical characteristics
| Characteristics | n (%) |
|---|---|
| Total eligible patients enrolled | 17 (100) |
| Enrollment time | September 11, 2013–April 10, 2017 |
| Age at study entry, median (range), yr | 57 (35–79) |
| Duration from diagnosis to study entry, mo | |
| Median (range) | 2.0 (0.6–5.1) |
| Missing | 1 |
| Gender | |
| Female | 6 (35) |
| Male | 11 (65) |
| Race or ethnicity | |
| White | 6 (35) |
| Hispanic | 8 (47) |
| Asian | 3 (18) |
| Tumor clinical stage at study entry by ERUS | |
| II | 6 (35) |
| III | 11 (65) |
| Distance of tumor to anal verge, median (range), cm | 8 (5–14) |
| Characteristics | n (%) |
|---|---|
| Total eligible patients enrolled | 17 (100) |
| Enrollment time | September 11, 2013–April 10, 2017 |
| Age at study entry, median (range), yr | 57 (35–79) |
| Duration from diagnosis to study entry, mo | |
| Median (range) | 2.0 (0.6–5.1) |
| Missing | 1 |
| Gender | |
| Female | 6 (35) |
| Male | 11 (65) |
| Race or ethnicity | |
| White | 6 (35) |
| Hispanic | 8 (47) |
| Asian | 3 (18) |
| Tumor clinical stage at study entry by ERUS | |
| II | 6 (35) |
| III | 11 (65) |
| Distance of tumor to anal verge, median (range), cm | 8 (5–14) |
Abbreviation: ERUS, endorectal ultrasound.
Study treatment and outcomes
| Outcome | n (%) |
|---|---|
| Tumor stage after complete neoadjuvant therapy by ERUS | |
| 0 | 1 (7) |
| I | 2 (13) |
| II | 7 (47) |
| III | 5 (33) |
| Surgery procedure performed | |
| APR | 2 (12) |
| LAR | 15 (88) |
| NAR score, as defined from NSABP R‐04 | |
| Low (<8) | 1 (6) |
| Intermediate (8–16) | 12 (71) |
| High (>16) | 4 (24) |
| Median (range) | 14.98 (0.94–41.62) |
| NAR quantiles | |
| 1st (<8.5) | 7 (41) |
| 2nd and 3rd (8.5–15.0) | 6 (35) |
| 4th (>15.0) | 4 (24) |
| Overall survival, median (95% CI), mo | Not reached |
| Disease‐free survival, median (95% CI), mo | Not reached |
| Duration follow‐up, median (range), mo | 44.8 (27.4–63.9) |
| Outcome | n (%) |
|---|---|
| Tumor stage after complete neoadjuvant therapy by ERUS | |
| 0 | 1 (7) |
| I | 2 (13) |
| II | 7 (47) |
| III | 5 (33) |
| Surgery procedure performed | |
| APR | 2 (12) |
| LAR | 15 (88) |
| NAR score, as defined from NSABP R‐04 | |
| Low (<8) | 1 (6) |
| Intermediate (8–16) | 12 (71) |
| High (>16) | 4 (24) |
| Median (range) | 14.98 (0.94–41.62) |
| NAR quantiles | |
| 1st (<8.5) | 7 (41) |
| 2nd and 3rd (8.5–15.0) | 6 (35) |
| 4th (>15.0) | 4 (24) |
| Overall survival, median (95% CI), mo | Not reached |
| Disease‐free survival, median (95% CI), mo | Not reached |
| Duration follow‐up, median (range), mo | 44.8 (27.4–63.9) |
Abbreviations: APR, abdominal perineal resection; CI, confidence interval; ERUS, endorectal ultrasound; LAR, low anterior resection; NAR, neoadjuvant rectal.
Study treatment and outcomes
| Outcome | n (%) |
|---|---|
| Tumor stage after complete neoadjuvant therapy by ERUS | |
| 0 | 1 (7) |
| I | 2 (13) |
| II | 7 (47) |
| III | 5 (33) |
| Surgery procedure performed | |
| APR | 2 (12) |
| LAR | 15 (88) |
| NAR score, as defined from NSABP R‐04 | |
| Low (<8) | 1 (6) |
| Intermediate (8–16) | 12 (71) |
| High (>16) | 4 (24) |
| Median (range) | 14.98 (0.94–41.62) |
| NAR quantiles | |
| 1st (<8.5) | 7 (41) |
| 2nd and 3rd (8.5–15.0) | 6 (35) |
| 4th (>15.0) | 4 (24) |
| Overall survival, median (95% CI), mo | Not reached |
| Disease‐free survival, median (95% CI), mo | Not reached |
| Duration follow‐up, median (range), mo | 44.8 (27.4–63.9) |
| Outcome | n (%) |
|---|---|
| Tumor stage after complete neoadjuvant therapy by ERUS | |
| 0 | 1 (7) |
| I | 2 (13) |
| II | 7 (47) |
| III | 5 (33) |
| Surgery procedure performed | |
| APR | 2 (12) |
| LAR | 15 (88) |
| NAR score, as defined from NSABP R‐04 | |
| Low (<8) | 1 (6) |
| Intermediate (8–16) | 12 (71) |
| High (>16) | 4 (24) |
| Median (range) | 14.98 (0.94–41.62) |
| NAR quantiles | |
| 1st (<8.5) | 7 (41) |
| 2nd and 3rd (8.5–15.0) | 6 (35) |
| 4th (>15.0) | 4 (24) |
| Overall survival, median (95% CI), mo | Not reached |
| Disease‐free survival, median (95% CI), mo | Not reached |
| Duration follow‐up, median (range), mo | 44.8 (27.4–63.9) |
Abbreviations: APR, abdominal perineal resection; CI, confidence interval; ERUS, endorectal ultrasound; LAR, low anterior resection; NAR, neoadjuvant rectal.
ClinicalTrials.gov Identifier: NCT01871571
Sponsor: University of Southern California
Principal Investigator: Afsaneh Barzi
IRB Approved: Yes
