nab-Paclitaxel for the treatment of pancreatic cancer

Abstract

Background

Nanoparticle albumin-bound paclitaxel (nab-P) plus gemcitabine (Gem) became a standard treatment option for metastatic pancreatic cancer (MPC) following positive results from a global phase III trial (MPACT). A large number of studies have now published results on the use of nab-P/Gem to treat advanced and early-stage disease, warranting a comprehensive review. The main goal of this systematic review is to summarize the efficacy and safety data of nab-P/Gem for the treatment of pancreatic cancer (PC).

Methods

This systematic review includes results from studies that either published results in a peer-reviewed journal or presented the results at a major oncology conference.

Results

Sixty-two studies were included (50 in the advanced/metastatic setting and 12 in the locally advanced setting). Most studies on the treatment of MPC were exclusively first line (33/50). Nevertheless, the studies in this review comprised a broad spectrum of patients, including those <65 and ≥65 years of age and those with a Karnofsky performance status of 70–100. Median overall survival (OS) in studies of nab-P/Gem in the advanced/metastatic setting ranged from 8.7 to 13.5 months. In addition, 15 studies of patients with advanced/metastatic PC examined nab-P/Gem as a backbone on which to add a variety of agents, including cancer stem cell inhibitors, stromal disrupting agents, and immune-modulating agents (median OS, 6.9–17 months). Ongoing trials are investigating nab-P/Gem with or without other agents across disease settings.

Discussion

Studies conducted after MPACT have demonstrated that nab-P/Gem is an effective regimen for the first-line treatment of MPC for a wide range of patients. Regimens using nab-P/Gem as a backbone on which to combine additional agents are being studied actively, particularly in the advanced disease setting. Ongoing studies will yield valuable insights on the utility of nab-P–containing regimens to improve patient outcomes in PC in both earlier-stage and advanced disease.

Keywords:

• pancreatic cancer
• nab-paclitaxel
• metastatic
• neoadjuvant
• systematic review

Introduction

More than 50,000 new pancreatic cancer (PC) cases and >40,000 cancer-related mortalities due to PC are expected in the USA in 2016.^1,2 The 5-year survival rate for all stages of PC combined is 8%. Although those with resectable disease have a more favorable prognosis (5-year survival ≈29%), ≈52% of patients are diagnosed with metastatic disease, which confers a less favorable outlook (5-year survival ≈3%).² Since the approval of gemcitabine (Gem) in 1997, no phase III trial in advanced/metastatic disease had demonstrated a clinically and statistically significant improvement in overall survival (OS) over Gem alone³ until recently. The treatment landscape for metastatic disease has evolved to include 2 key regimens: folinic acid, 5-fluorouracil, irinotecan, and oxaliplatin (FOLFIRINOX) and nanoparticle albumin-bound paclitaxel (nab-P) plus Gem (nab-P/Gem). The FOLFIRINOX regimen was approved based on a French multicenter phase II/III trial that reported significant improvements in OS with FOLFIRINOX versus Gem (median, 11.1 vs 6.8 months; hazard ratio [HR], 0.57; P<0.001), but significant adverse events were also observed.⁴ The nab-P/Gem regimen was approved in many countries after the phase III MPACT trial demonstrated that the addition of nab^®-P (Abraxane^®; Celgene Corporation, Summit, NJ, USA) to Gem improved OS versus Gem (median, 8.7 vs 6.6 months; HR, 0.72; P<0.001).⁵ Currently, the National Comprehensive Cancer Network (NCCN) and European Society for Medical Oncology (ESMO) recommend treatment with FOLFIRINOX or nab-P/Gem as standards of care for patients with metastatic pancreatic cancer (MPC).^6,7 Age, performance status (PS), and other clinical factors are considered when deciding which regimen to use; Gem monotherapy is currently reserved for patients ineligible to receive combination chemotherapy.⁶

nab-P/Gem and FOLFIRINOX have not been approved for earlier-stage disease; however, numerous trials are exploring their utility. The NCCN recommends chemotherapy for unresectable locally advanced PC (LAPC) and chemoradiation for selected patients, preferably after induction chemotherapy for tumor control.⁶ Currently, no clear evidence exists to support the use of nab-P/Gem over FOLFIRINOX or vice versa, and several trials are investigating their efficacy and safety.⁸

A population-based study of >3,000 patients showed that nab-P/Gem is the most commonly used chemotherapy regimen for the first-line treatment of MPC in the USA,⁹ possibly due to the toxicity profile of FOLFIRINOX, which limits its use to younger/fitter patients. The extensive use of nab-P/Gem in both academic and community settings coupled with >100 current and active clinical trials in PC warrants a comprehensive review of clinical data to gain a better understanding of how this regimen is being used for the treatment of PC and associated outcomes. The overall goal of this review is to summarize recent data regarding the safety and efficacy of regimens that include nab-P/Gem for patients with PC.

Methods

The search terms “nab-paclitaxel and (pancreatic or pancreas)” were entered in PubMed to retrieve publications from January 1, 2011 to June 30, 2016. Abstracts from the annual meetings of the American Society of Clinical Oncology (ASCO) 2011–2016, the Gastrointestinal Cancers Symposium (ASCO GI) 2011–2016, the European Cancer Organisation/ESMO 2011–2015, the ESMO World Congress on Gastrointestinal Cancer 2015 and 2016, and the Italian Association of Medical Oncology (2014) were searched using the term “nab-paclitaxel.” Clinical trials and institutional analyses of nab-P in all stages of PC were included. Duplicates, electronic abstracts, case studies, cost studies, meta-analyses, and studies of the effects of eligibility criteria were excluded. The website www.clinicaltrials.gov was searched using the terms “nab-paclitaxel” OR “Abraxane” AND “pancreatic” AND “adenocarcinoma” to identify ongoing trials without results; only open, active, phase II–III trials with a sample size ≥100 were included.

Results

Studies of nab-P in advanced/metastatic PC

Fifty studies evaluating nab-P in MPC were retrieved (Figure 1; Table 1). Approximately one-half were retrospective analyses. MPACT was the only phase III study, and all other prospective trials were phase I or II. Two-thirds of studies evaluated nab-P in the first-line setting, and approximately one-third of those studies assessed nab-P/Gem with an additional agent. nab-P was most often evaluated at a dose of 125 mg/m², which was given the first 3 of 4 weeks (qw 3/4). The tables in this systematic review cover MPC (Tables 1–3), neoadjuvant treatment or locally advanced disease (Table 4), and ongoing trials in all settings (Table 5).

Table 1 Characteristics of advanced/MPC studies (n=50)

Characteristic	Number of studies (%)
Study design	1 (2)
Prospective	9 (18)
Pilot	6 (12)
Phase I	10 (20)
Phase I/II	1 (2)
Phase II	23 (46)
Phase III
Retrospective/institutional experience
First-line only nab-P
Regimen	33 (66)
nab-P/Gem	29 (58)
nab-P/Gem + other agent	15 (30)
nab-P + other agent	3 (6)
nab-P monotherapy	3 (6)

Abbreviations: Gem, gemcitabine; MPC, metastatic pancreatic cancer; nab-P, nanoparticle albumin-bound paclitaxel.

Table 2 Overall survival (OS) with first-line nab-P/Gem in studies of ≥45 patients with metastatic pancreatic cancer

First author, year	Type of study	Agent(s)a	n	Age, median, years	PS	Median OS (95% CI), months	P value
Von Hoff, 2011^Citation10	Ph I/II	nab-Pb/Gem	67	61c	ECOG 0–1	12.2c	NR
Cartwright, 2014^Citation11	Retro	nab-Pd/Gem	189	Gem-based regimens: 70	Gem-based regimens:	10.2	NR
		Gem + other chemo	1,567		KPS <70, 7%	7.0
		FOLFIRINOX	666	60	KPS <70, 1%	11.2
Santoni, 2014^Citation12	Retro	nab-Pd/Gem	41	66	NR	11.6	NR
		Gem	159			5.5
		Gem + cisplatin/oxaliplatin	234			7.5
		Gem + capecitabine	43			9.1
		FOLFIRINOX	101			13.0
Krishna, 2015^Citation13	Retro	nab-P/Geme	49	65	ECOG 0–1	11.1 (5.3–not reached)	NA
MPACT	Ph III	nab-P/Gem	431	62	KPS <80, ≈7%	8.7 (7.9–9.7)	<0.001
Goldstein, 2015^Citation5		Gem	430	63	KPS <80, 8%	6.6 (6.0–7.2)
Giordano, 2015^Citation50	Retro	nab-Pf/Gem	208	67	ECOG PS 2, 17.8%	11 (8.8–13.2)	NA
Shen, 2016^Citation17	Ph II	nab-P/Gem	83	57	KPS 70–80, 30%	9.2 (5.29–7.16)	NA
Hammel, 2016^Citation15	Ph II	nab-P/Gem	39	65.3	ECOG 2, 15.4%	9.2 (6.0–13.6)	NR
		nab-P + sLV5FU2	75	66.2	ECOG 2, 16.0%	11.4 (8.8–16.6)

Notes:

a nab-P at 125 mg/m² the first 3 of 4 weeks (qw 3/4) unless otherwise indicated.

b nab-P at 100, 125, or 150 mg/m² qw 3/4.

c For nab-P 125 mg/m² qw 3/4 (n=44).

d Dose and schedule of nab-P not reported.

e nab-P at 125 mg/m² and Gem at 1,000 mg/m² both given q2w. Supportive care also included dexamethasone 12 mg 30 min prior to chemotherapy administration.

F nab-P at 100 or 125 mg/m² qw 3/4.

Abbreviations: chemo, chemotherapy; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; FOLFIRINOX, folinic acid, 5-fluorouracil, irinotecan, and oxaliplatin; Gem, gemcitabine; KPS, Karnofsky performance status; NA, not applicable; NR, not reported; nab-P, nanoparticle albumin-bound paclitaxel; Ph, phase; PS, performance status; Retro, retrospective; sLV5FU2, simplified leucovorin and 5-fluorouracil regimen.

Table 3 Studies of nab-P/Gem + another agent for advanced/metastatic pancreatic cancer (no cutoff based on N)

First author, year	Type of study	Line of Tx	Agent combined with nab-Pa/Gem	N	MPC, %	Age, median, years	PS	Median OS (95% CI), months
Cohen, 2016^Citation56	Ph Ib	1st	Erlotinibb	19	63	63	ECOG 0–1	9.3 (3.3–15.4)
Ko, 2012^Citation57	Ph I	1st	Capecitabinec	15	100	62	ECOG 0–2	7.5 (NR)
De Jesus-Acosta, 2014^Citation58	Ph II	1st	Vismodegib added in cycle 2	59	100	60	ECOG 0–1	10 (7.3–11)
ALPINE	Ph Ib	1st	Tarextumab	40	100	63	ECOG 0–1	11.6
O’Reilly, 2015^Citation59
Hidalgo, 2016^Citation60	Ph Ib	1st	Demcizumab Gem + demcizumab (no nab-P)	56	70	65	NR	10.1 (6.5–16.2) NR
Hingorani, 2016^Citation61–Citation63	Ph II	1st	PEGPH20	74	100	NR	NR	12 (high-HA population)
			nab-P/Gem only	61				9 (high-HA population)
O’Reilly, 2016^Citation64	Ph I	1st	Necuparanib	27	100	63 (mean)	ECOG 0–1	13.1 (4.0–16.6) for patients who completed ≥1 dose
			Gem + necuparanib (no nab-P)	12				10.4 (6.1–21.8) for patients who completed ≥1 dose
Bhattacharyya, 2015^Citation65	Inst.	1st	VT-122CM nab-P/Gem only	20 17	65 76	62 60	Mean ECOG 1.9 Mean ECOG 2.1	17.0 9.3 (P<0.001)
Mahipal, 2015^Citation66	Ph I	1st	Enzalutamide	8	100	64	ECOG 1	NR
Reni, 2014^Citation67	Ph Ib	1st	Capecitabine + cisplatind	24	NR	63	KPS≤80, 13%	NR
Sigal, 2013^Citation68	Ph II	1st	2-O, 3-O desulfated heparin (ODSH)	10	NR	66	ECOG 0–1	NR
RAINIER	Ph II	1st	Apatorsen	66	100	67	ECOG 0–1	5.3 (3.2–7.2)
Ko, 2016^Citation69			nab-P/Gem only	66	100	66	ECOG 0–1	6.9 (P=NS)
El-Rayes, 2016^Citation70	Ph Ib	≤2nd	BBI-608	37	100	63	ECOG 0–1	NR
Bahary, 2016^Citation71	Ph Ib	1st	Indoximod	15	100	68	KPS≥70	NR
Borad, 2016^Citation72	Ph 1	1st	Evofosfamide	19	89	62	ECOG 0–1	14.2 (8.5–19.4)

Notes:

a nab-P at 125 mg/m² the first 3 of 4 weeks (qw 3/4) unless otherwise indicated.

b nab-P at 75, 100, or 125 mg/m² qw 3/4.

c Dose escalation of nab-P from 100 to 150 mg/m² on day 4 of a 14-day cycle.

d nab-P at 100–150 mg/m² on days 1 and 14 every 4 weeks.

Abbreviations: CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; Gem, gemcitabine; HA, hyaluronan; Inst, institutional analysis; KPS, Karnofsky performance status; MPC, metastatic pancreatic cancer; NR, not reported; nab-P, nanoparticle albumin-bound paclitaxel; NS, not statistically significant; OS, overall survival; Ph, phase; PS, performance status; Tx, treatment.

Table 4 Locally advanced and/or earlier-stage pancreatic cancer studies of ≥15 patients that include treatment with nab-P/Gem

First author, year	Type of study	Regimena	N	Stage	Age, median, years	Response data	Resection rate in all patients/in patients who underwent resection
							R0	R1
Sueyoshi, 2015^Citation51	Ph I	nab-Pb/Gem + radiation	15	Unresectable LAPC	63	PR=13% SD=67% PD=7%	NA	NA
Dean, 2016^Citation52	Retro	nab-P/Gem → 5-FU CRT	42	Unresectable LAPC	66	pCR=33%	7%/38%	12%/63%
Idrees, 2016^Citation33	Retro	nab-Pc/Gem	26	BL resectable (77%) and LAPC (23%)	NR	pCR=15%	NR/86% (not given for each group)	NR
		FOLFIRINOX	59	BL resectable (63%) and LAPC (37%)	NR	pCR=5%		NR
Peterson, 2016^Citation53	Retro	nab-P/Gem	20	BL resectable (70%) and unresectable (30%); patients ineligible for FOLFIRINOX	69	PR=20%	20%/67%	NA
NEOPAX, Van Laethem, 2016^Citation54	Ph 0	nab-P/Gem	23	Unresectable and borderline resectable	63	PR=35% pCR=0	30%/NR	26%/NR
GAIN-1; Sliesoraitis, 2014^Citation55	Ph II	nab-Pd/Gem	10	Resectable/borderline resectable	68		60%/75%	20%/25%
		Non-neoadjuvant historic controls	22		67		77%/NR	9%/NR
Alvarez, 2013^Citation40	NR	nab-P/Gem	16	Resectable, 44%; borderline resectable, 56%	58	PR by PET, 50%; no objective responses; 1 complete pathological response, 6 GRT-1, 1 GRT-2, 2 GRT-3	69%/92%	6%/8%
GAP; Barbour, 2015^Citation39	Ph II	nab-P/Gem	41	Resectable	65	Pancreatic resection rate, 73%	1-mm margin: 37%/52% 0-mm margin: 61%/86%	1-mm margin: 34%/48% 0-mm margin: 10%/14%
MacKenzie, 2013^Citation38	Ph II	nab-P/Gem	25	Resectable	65	RECIST  PR=36%  SD=18%  PD=8%	80%/95%	4%/5%

Notes:

a nab-P at 125 mg/m² the first 3 of 4 weeks (qw 3/4) unless otherwise indicated.

b nab-P at 50–125 mg/m² qw 3/4.

c Dose and schedule of nab-P not reported.

d nab-P at 100 mg/m² qw 3/4.

Abbreviations: 5-FU, 5-fluorouracil; BL, baseline; CRT, chemoradiation therapy; FOLFIRINOX, folinic acid, 5-fluorouracil, irinotecan, and oxaliplatin; Gem, gemcitabine; GRT, grade of residual tumor; LAPC, locally advanced pancreatic cancer; NA, not applicable; NR, not reported; nab-P, nanoparticle albumin-bound paclitaxel; pCR, pathological complete response; PET, positron emission tomography; Ph, phase; PD, progressive disease; PR, partial response; RECIST, Response Evaluation Criteria In Solid Tumors; Retro, retrospective; SD, stable disease.

Table 5 Selected ongoing phase II/III trials (N≥100) of nab-P/Gem ± other agents in pancreatic adenocarcinoma

Trial	Phase	Planned N	Patient population or stage of disease	Regimen	Planned primary endpoints
Metastatic or advanced stage nab-P/Gem only
QOLINPAC, NCT02106884^Citation73	II	110	Unresectable LAPC or metastatic	First-line nab-P/Gem vs Gem	Deterioration-free QOL using EORTC QLQ-C30
ALPACA, NCT02564146^Citation74	II	325	Metastatic	First-line: induction with nab-P/Gem → nab-P/Gem vs induction with nab-P/Gem → nab-P/Gem or alternating Gem monotherapy and nab-P/Gem	OS
nab-P/Gem + other
NCT02101021^Citation75	III	430	Metastatic	First-line nab-P/Gem + momelotinib vs nab-P/Gem	DLT, OS
NCT02715804^Citation76	III	420	Metastatic	First-line nab-P/Gem + PEGPH20 vs nab-P/Gem + placebo	PFS, OS
RESOLVE, NCT02436668^Citation46	II/III	326	Metastatic	First-line nab-P/Gem + Ibrutinib vs nab-P/Gem + placebo	PFS
CARRIE, NCT02399137^Citation36	II	260	Metastatic	First-line nab-P/Gem + MM-141 vs nab-P/Gem + placebo	PFS
YOSEMITE, NCT02289898^Citation35	II	201	Metastatic	First-line nab-P/Gem + placebo vs nab-P/Gem + demcizumab + placebo (truncated course of demcizumab) vs nab-P/Gem + demcizumab	PFS
NCT02551991^Citation77	II	168	Metastatic	First-line nab-P/Gem vs nal-IRI + 5-FU + folinic acid vs nal-IRI + 5-FU + folinic acid + oxaliplatin	PFS
FIRGEMAX, NCT02827201^Citation78	II	124	Metastatic	First-line nab-P/Gem alternating with FOLFIRI.3 vs nab-P/Gem	PFS at 6 months
SEQUENCE, NCT02504333^Citation79	I/II	180	Metastatic	First-line nab-P/Gem → recommended dose of modified FOLFOX from phase I	Phase I: safety, DLT Phase II: OS at 12 months
PACT-19, NCT01730222^Citation80	I/II	134	Advanced	Phase II: first-line nab-P RP2D + Gem 800 mg/m^2 + cisplatin 30 mg/m^2 + cape 1,250 mg/m^2 q2w every 4 weeks vs nab-P 125 mg/m^2 + Gem 1,000 mg/m^2 qw 3/4	Phase I: DLT Phase II: PFS for stage IV, resectability rate for stage III
NabucCO, NCT02109341^Citation81	I/II	114	Metastatic	First-line nab-P + FOLFIRI or nab-P + FOLFOX	MTD, DLTs, ORR
NCT02194829^Citation82	I/II	133	Advanced	First-line nab-P/Gem ± MK-1775	Phase I: MTD Phase II: PFS
Resectable or locally advanced nab-P/Gem only
LAPACT, NCT02301143^Citation44,Citation83	II	110	Untreated LAPC	nab-P/Gem	Time to treatment failure
APACT, NCT01964430^Citation45	III	800	Resected	Adjuvant nab-P/Gem vs Gem	DFS
NEONAX, NCT02047513^Citation47	II	166	Resectable	Neoadjuvant and adjuvant vs only adjuvant nab-P/Gem	Time to DFS
S1505, NCT02562716^Citation84	II	112	Resectable	Neoadjuvant nab-P/Gem vs mFOLFIRINOX	OS
NCT02506842^Citation48	III	300	Resected	Second-line adjuvant nab-P 100 mg/m^2 + Gem 1,000 mg/m^2 vs oxaliplatin + folinic acid + 5-FU	OS
NCT02243007^Citation42	II	112	Resectable	Neoadjuvant FOLFIRINOX vs nab-P/Gem	OS at 18 months
nab-P/Gem + other
NEOLAP, NCT02125136^Citation41	II	168	Untreated unresectable or borderline resectable LAPC	Neoadjuvant nab-P/Gem vs nab-P/Gem followed by FOLFIRINOX	Conversion rate to resection
“Personalized Medicine,” NCT01726582^Citation85	II	120	Resectable and borderline resectable	nab-P/Gem ± subsequent CRT with Gem or cape as neoadjuvant or adjuvant therapy vs other chemotherapies in similar settings vs CRT with Gem or cape in similar settings	Resectability rate
SCALOP-2, NCT02024009^Citation86	I/II	289	LAPC	Induction nab-P/Gem → nab-P/Gem + RT → cape + RT ± nelfinavir vs 6 cycles of nab-P/Gem	OS, PFS

Abbreviations: 5-FU, 5-fluorouracil; cape, capecitabine; CRT, chemoradiation therapy; DFS, disease-free survival; DLT, dose-limiting toxicity; EORTC, European Organisation for Research and Treatment of Cancer; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; FOLFIRINOX, folinic acid, 5-fluorouracil, irinotecan, and oxaliplatin; FOLFOX, folinic acid, 5-fluorouracil, and oxaliplatin; Gem, gemcitabine; LAPC, locally advanced pancreatic cancer; MTD, maximum tolerated dose; nal-IRI, nanoliposomal irinotecan; nab-P, nanoparticle albumin-bound paclitaxel; OS, overall survival; PFS, progression-free survival; QOL, quality of life; qw 3/4, first 3 of 4 weeks; RP2D, recommended phase II dose; RT, radiotherapy.

Figure 1 Schematic of method for systematically selecting studies for inclusion in the database.

Abbreviations: LAPC, locally advanced pancreatic cancer; nab-P, nanoparticle albumin-bound paclitaxel; PC, pancreatic cancer.

nab-P/Gem in MPC

Ten studies reported the median OS for first-line nab-P/Gem in patients with advanced PC;^5,10–18 Table 2 lists 8 of these studies with a population >45 patients. The most commonly used dose and schedule were those used in the MPACT trial: nab-P 125 mg/m² plus Gem 1,000 mg/m² administered on a qw 3/4 schedule.^5,14,19,20 Patients treated with this dose and schedule experienced a median OS ranging from 8.7 to 13.5 months^5,18 and 1-year survival ranging from 35% to 62%.^5,14,19,20 Most prospective studies evaluating this dose and schedule were single-arm trials.

nab-P/Gem in MPC – age

It may be expected that younger patients would experience longer survival and improved tolerability compared with older patients. However, most studies, including MPACT, suggest that older patients benefit from nab-P/Gem in terms of efficacy without increased risk of toxicity. Approximately 40% of patients enrolled in MPACT were >65 years.⁵ Median OS was 9.6 and 7.7 months for patients <65 and ≥65 years, respectively, and the toxicity profiles were similar between age groups.⁵ The combination in MPACT demonstrated significant OS benefit over Gem alone in both age groups: <65 years (HR, 0.65; P<0.001) and ≥65 years (HR, 0.80; P=0.048).

A study (N=37) including patients treated with first-line or ≥ second-line nab-P/Gem for MPC showed that OS was not significantly different between patients ≥66 years and those <66 years of age (median, 10.5 vs 9 months; P=0.49).²¹ Similarly, a large Italian database review of patients (N=208) with advanced PC treated with nab-P/Gem demonstrated that age (≥75 vs <75 years) was not significantly associated with efficacy or toxicity with respect to median OS (11.4 vs 11 months; P=0.86), disease control rate (69% vs 61%; P=0.64), grade 3/4 neutropenia (25% vs 28%), and neurotoxicity (9% vs 12%).²² Additionally, an exploratory analysis from MPACT showed that the percentages of patients requiring nab-P dose reductions were similar between age groups (42% for patients ≥65 years vs 40% for patients <65 years).²³

nab-P/Gem in MPC – PS

Data on whether patients with a better PS might receive greater benefit from nab-P/Gem than patients with a poorer PS are inconclusive; however, similar to the literature in older patients, several studies suggest that less fit patients receive meaningful benefit from the regimen. Stratification of the MPACT population by Karnofsky PS (KPS) demonstrated significantly better OS in the fitter (KPS 90–100) versus less fit (KPS 70–80) group in the combination arm (median, 9.7 vs 7.6 months; HR, 0.76; P=0.009) and the Gem arm (median, 7.9 vs 4.3 months; HR, 0.57; P<0.001).⁵ In the KPS 70–80 subpopulation, nab-P/Gem extended median OS by >3 months compared with Gem alone (7.6 vs 4.3 months; HR, 0.59; P<0.001).

A small phase I/II trial examined the effect of nab-P/Gem in patients with an Eastern Cooperative Oncology Group (ECOG) PS of 2.²⁴ The results of the phase I portion suggest that these patients were able to receive the standard dose of nab-P/Gem; the relative dose intensity was 100% in 6 patients who received nab-P 125 mg/m² plus Gem 1,000 mg/m² qw 3/4.

In a retrospective analysis of 39 patients with unresectable LAPC or MPC treated with nab-P/Gem,²⁵ patients with an ECOG PS of 1 survived longer than patients with an ECOG PS of 2 (median OS, 15 vs 7 months; P=0.032).²⁵ Similarly, the previously mentioned Italian retrospective analysis of patients with advanced PC (N=208) treated with nab-P/Gem showed a numerically shorter OS in the ECOG PS 2 versus ECOG PS 0–1 group (median, 8.7 vs 11.2 months; P=0.07), but the difference was not significant.²² In addition, toxicities did not appear to be influenced by PS, because similar percentages of patients with PS 0–1 and PS 2 developed neutropenia (31% and 34%, respectively) and neurotoxicity (17% in each group). Collectively, these studies suggest that, although PS may affect OS, nab-P/Gem seems to be effective regardless of PS.

nab-P/Gem in MPC – real-world comparative effectiveness studies

Although clinical trials comparing nab-P/Gem with FOLFIRINOX for the treatment of PC have not yet reported results, retrospective analyses have explored these standard-of-care regimens with one another and/or Gem for the treatment of MPC.^{11,12,26–29} One study reported a median OS of 10.2 months with nab-P/Gem (n=189) versus 11.2 months with FOLFIRINOX (n=666) and 7 months for Gem combined with other chemotherapies (n=1,567).¹¹ Similar results were reported from another retrospective analysis: median OS of 11.6 months with nab-P/Gem (n=41) versus 13 months with FOLFIRINOX (n=101) and 7.5–9.1 months for Gem plus other chemotherapies (n=277).¹² A real-world analysis based on electronic medical records of patients (N=202) receiving first-line treatment for advanced PC demonstrated similar comparative effectiveness for nab-P/Gem versus FOLFIRINOX (database persistence [proxy for OS], median, 8.6 months in both groups), despite patients in the FOLFIRINOX group being significantly younger.²⁷ In addition, a retrospective analysis (N=150) of patients treated at 5 cancer centers in British Columbia, Canada, found that both nab-P/Gem and FOLFIRINOX produced similar outcomes and demonstrated longer OS versus Gem alone as treatment for unresectable PC (median, 11.6 and 11.2 vs 4.1 months, respectively; P<0.001 and P=0.039).²⁹ Patients who received FOLFIRINOX were younger (median age, 61 vs 70 years) and fitter (ECOG PS≤1, 91% vs 54%) than those who received nab-P/Gem.²⁹ Collectively, the OS with nab-P/Gem observed in MPACT was consistent with the OS observed in real-world observational data sets, and nab-P/Gem was comparable in effectiveness to FOLFIRINOX.

Subsequent therapies after first-line nab-P/Gem in MPC

Many recent analyses have examined the use of second-line therapies after nab-P/Gem.^27,30–33 Patients in MPACT who received second-line therapy (n=170) after nab-P/Gem experienced a numerically longer median OS than those who did not (n=250; median total OS, 12.8 and 6.3 months, respectively).³⁰ The longest total OS values were observed in patients who received first-line nab-P/Gem followed by fluoropyrimidine-containing second-line regimens (n=132; median, 13.5 months); a small number (n=18) received second-line FOLFIRINOX and experienced a median total OS of 15.7 months.³⁰ Another retrospective analysis from the previously described Italian registry (N=250) demonstrated similar findings, that is, a median OS of 13.5 months in patients who received second-line treatment after first-line nab-P/Gem (n=122).³¹ More specifically, patients who received second-line FOLFOX/XELOX (n=56), FOLFIRI (n=24), and FOLFIRINOX (n=22) had median total OS values of 12.8, 13.2, and 13.8 months, respectively.³¹ Consistent findings have been observed in many other analyses, and the totality of data suggests that first-line nab-P/Gem followed by second-line therapy, particularly with regimens that contain a fluoropyrimidine, is feasible and beneficial to patients with advanced PC.^27,30–33

Future directions

Future directions for nab-P/Gem include studies in which the regimen has been used as a backbone therapy (ie, with another agent) in MPC (Table 3) and as a doublet in locally advanced pancreatic cancer (Table 4). Table 5 displays a list of selected ongoing trials of nab-P/Gem with or without other agents as treatment for metastatic, locally advanced, and resectable disease.

nab-P/Gem as a backbone regimen in MPC (studies with results)

Because nab-P/Gem has demonstrated survival comparable to that with FOLFIRINOX and a more favorable toxicity profile, this regimen is commonly used as a chemotherapy backbone for other agents (Table 3). Agents combined with nab-P/Gem are diverse and include cancer stem cell inhibitors (demcizumab, vismodegib, tarextumab, and BBI-608), those with potential immune-modulating activities (indoximod), those directed against tumor stroma (PEGPH20 and 2-0, 3-0 desulfated heparin), chemotherapies (capecitabine ± cisplatin), hormone therapy (enzalutamide), and others (erlotinib and apatorsen). In 15 studies of patients with MPC treated with nab-P/Gem combined with other agents (including 10 phase I trials), the median OS ranged from 6.9 to 17 months.

nab-P/Gem as a backbone regimen in MPC (studies without results)

Thirty ongoing phase II and III trials of nab-P in PC with a sample size of ≥100 were identified, including 16 MPC trials (all first line); most included an additional agent (Table 5). For example, the phase II/III RESOLVE trial (N=326) is evaluating nab-P/Gem, with or without the Bruton tyrosine kinase inhibitor ibrutinib, as first-line treatment of MPC.³⁴ Based on promising results from phase I/II trials (Table 3), a phase III trial (N=420) is investigating PEGPH20 in combination with nab-P/Gem in patients with high levels of hyaluronan, and demcizumab with nab-P/Gem is being evaluated in the phase II YOSEMITE trial (N=201).³⁵ Another noteworthy ongoing trial is a phase II study (N=260) of nab-P/Gem plus istiratumab (MM-141; a bispecific antibody against ErbB3 and insulin-like growth factor-1 [IGF-1] receptor) for the first-line treatment of patients with MPC and high serum levels of free IGF-1.³⁶ Finally, whether the combination of nab-P/Gem with checkpoint inhibitors will be an effective strategy for PC is an important question, because checkpoint inhibitors have recently provided break-through treatment options for several tumor types and are currently being explored in a number of PC trials. Data on such combinations (eg, nab-P/Gem and nivolumab)³⁷ are preliminary at this point.

Neoadjuvant trials for patients with resectable, borderline resectable, or LAPC (studies with results)

Several recent studies (n=12) examined neoadjuvant nab-P/Gem as a strategy for improving R0 resection rates in resectable tumors or converting borderline resectable tumors to resectable tumors. One of the main pathologic predictors of survival after surgery is resection margin status; a negative resection margin (R0) is associated with better prognosis compared with a positive margin. Eight of the 12 studies had a total enrollment of ≥15 patients (Table 4). Noteworthy among these is a pilot phase II study in which patients with resectable PC (N=25) were treated with neoadjuvant nab-P/Gem for 3 cycles.³⁸ Surgical resection was possible in 84% of patients and resulted in R0 resection in 95% of resected cases, or 80% of the intention-to-treat population.³⁸ The phase II GAP study also evaluated neoadjuvant nab-P/Gem for 2 cycles in patients with resectable PC (N=41).³⁹ After neoadjuvant treatment, 73% of the patients underwent pancreatic resection.³⁹ Similar results were reported from another trial of neoadjuvant nab-P/Gem (administered for 2 cycles) in patients with resectable or borderline resectable tumors (N=16).⁴⁰ Seventy-five percent of patients underwent surgery, and R0 resection was achieved in 69% of the intention-to-treat population – 92% of those who underwent surgery.

Neoadjuvant trials for patients with resectable, borderline resectable, or LAPC (studies without results)

The phase II NEOLAP trial (N=168) will examine the ability of neoadjuvant nab-P/Gem versus FOLFIRINOX to convert unresectable LAPC or borderline resectable tumors to resectable tumors (Table 5).⁴¹ Another phase II study (N=112) is comparing neoadjuvant nab-P/Gem versus FOLFIRINOX followed by resection in patients with potentially resectable tumors.⁴² The randomized phase II LAPACT study (N=110) is investigating time to treatment failure in patients with unresectable LAPC treated with nab-P/Gem.^43,44

Ongoing adjuvant trials for patients with resectable PC

The ongoing phase III APACT study is evaluating nab-P/Gem versus Gem monotherapy as adjuvant treatment in patients who have undergone macroscopic complete resection for non-MPC (Table 5).^45,46 Two other studies are also examining nab-P/Gem as adjuvant therapy: the phase II NEONAX study (N=166; nab-P/Gem as adjuvant only vs as neoadjuvant plus adjuvant)⁴⁷ and a second-line adjuvant phase III trial in patients who experienced disease relapse during Gem-based adjuvant therapy (N=300).⁴⁸

Discussion

Multiple studies have demonstrated that first-line treatment with nab-P/Gem improves survival in patients with MPC, with OS similar to or better than that observed in MPACT. These studies have helped to confirm the dose and schedule of nab-P 125 mg/m² plus Gem 1,000 mg/m² qw 3/4 as an effective and tolerable option for patients with MPC. Retrospective analyses of comparisons between nab-P/Gem and FOLFIRINOX suggested similar efficacy outcomes between the regimens, despite differences in patient populations; nab-P/Gem was used in a broader spectrum of patients.

Most studies demonstrated an OS benefit with nab-P/Gem regardless of age group; similarly, patients seem to derive substantial clinical benefit from nab-P/Gem regardless of PS. The demonstrated efficacy of first-line nab-P/Gem has led to a number of studies examining regimens afterward as second-line therapy.^27,30–33 These studies showed that second-line treatment after nab-P/Gem is feasible and that fluoropyrimidine-containing regimens, and not exclusively FOLFIRINOX, are appropriate options in this setting.

There are currently >100 ongoing trials (combined target enrollment >9,500 patients) assessing different nab-P regimens for the treatment of PC, and these studies will provide critical information regarding optimal combinations for specific patient populations.⁴⁹

Conclusion

In summary, nab-P/Gem is an effective and well-tolerated regimen for patients with PC. Ongoing trials will evaluate nab-P in all stages of PC. The combination of nab-P/Gem has become a standard of care for MPC and a backbone onto which novel therapies are added in ongoing trials. Future directions in this field will revolve around improving our understanding of PC, including its molecular biology, and identifying subsets of patients that may benefit from specific treatments.

Acknowledgments

Medical writing assistance was provided by John McGuire, PhD, MediTech Media, Ltd, funded by Celgene Corporation. The author is fully responsible for content and editorial decisions for this manuscript. The author is on the speaker’s bureau and is a consultant for Celgene Corporation.

Disclosure

The author reports no other conflicts of interest in this work.