Methods
In an online meeting held in December 2022, sixteen medical oncologists conducted
and went through a series of presentations on the management of mCRC. The presented
evidence was then further reviewed and condensed into a set of consensus statements.
The statements were primarily prepared by two of the authors, while two authors provided
additional inputs on the same. All the prepared consensus statements were then reviewed
in an in-person meeting in February 2023 by a group of 38 medical oncologists. Attempts
were made to involve clinicians from publicly funded hospitals and private healthcare
institutions which were high-volume centers for the management of CRC. Each statement
was assessed as Agree or Disagree with voting being performed anonymously. A statement
was considered as “Accepted” if greater than 70% of the group voted to agree with
the statement. In cases where a consensus was not met, statements were revised and
voted on again till a consensus was reached. The group evaluated each statement's
level of evidence and grade of recommendation as per IDSA-US Public Health Service
Grading System (USPHS). As per this system, the letters A to E signify the strength
of the recommendation for or against a preventive or therapeutic measure, while the
Roman numerals I to III indicate the quality of evidence supporting the recommendation.[10] All recommendations were graded as per this classification excepting those with
regard to radiological studies, which was voted upon as “recommended” or “not recommended”
only. Aspects with regard to radiological staging and radiotherapy were independently
assessed and separately modified based on inputs from a dedicated gastrointestinal
radiologist and radiation oncologist from Tata Memorial Hospital, Mumbai.
Consensus Statements
Accurate radiological assessment at baseline in suspected mCRC assumes utmost importance,
especially in patients with limited and potentially resectable metastases. Ruling
out the presence of multiple unresectable metastases in the liver or other organs
at baseline is the first step in planning for resection of colorectal liver metastases
(CRLMs). This step is crucial to avoid unnecessary resections in patients who otherwise
have widely disseminated multiorgan metastases. However, at the same time, accurate
baseline staging would justify major, curative liver resections as the 10-year survival
rate of mCRC patients with resectable liver metastases who undergo curative liver
resections approximately 40 to 50%.[11]
A well-conducted triphasic CECT of the thorax, abdomen, and pelvis as baseline imaging
has an accuracy of about 95% in the detection of distant metastases.[12]
[13]
[14] Its low cost (with the caveat of decreasing costs associated with the performance
of PET-CT as opposed to the CT scan of the thorax, abdomen, and pelvis), widespread
availability, and high anatomic resolution make it the imaging of choice.[15] However, a CT cannot accurately discriminate neoplastic disease from nonmalignant
changes like scars and inflammation (especially with coexisting liver parenchymal
disease) and may miss small tumours.[16]
The major advantages of a PET scan over and above a well-conducted triphasic CECT
TAP include total body coverage, including bone and a higher sensitivity in the detection
of hepatic and lung metastases.[17]
[18] FDG PET has an accuracy of 99% in the detection of liver metastases with a sensitivity
close to 100% and a specificity of 98%.[18] There are many studies that have shown that FDG PET changes the stage and alters
the management in up to one-third of the patients.[19]
The disadvantages of a PET scan include high background activity of liver, which may
make detection of very small liver metastases difficult as well as poor spatial resolution
of FDG PET, which occasionally makes surgical planning difficult.
In resource-constrained settings where a PET scan may not be available, a well-conducted
triphasic CECT of the abdomen is sufficient as a baseline scan to take surgical decisions
regarding metastasectomy. In scenarios where CT scans are of inadequate quality or
available CT plates do not convey the requisite information (despite a well-conducted
CECT), efforts must be made to obtain Digital Imaging and Communications in Medicine
(DICOM) images before considering an FDG PET CECT scan. However, if efforts to procure
adequate images are unlikely, an FDG PET CECT scan may be offered in light of its
superior sensitivity in picking up smaller metastases.
The general preference for a triphasic CECT scan of the abdomen as the imaging modality
of choice for evaluation of liver metastases from mCRC is not wholly necessary unless
a hyper vascular lesion suggestive of a neuroendocrine tumor (wherein a biphasic scan
comprising arterial and venous phases is adequate) or a cholangiocarcinoma/hepatocellular
carcinoma (wherein a triphasic scan comprising arterial, venous, and portal phases
is required) is suspected. A single-phase CT scan is adequate for staging of the liver
in patients with mCRC with liver metastases; however, surgeons may require mapping
of the arterial anatomy post-induction/neoadjuvant therapy and hence may consider
triphasic scans. CT scans are also widely available and cheaper than MRI in resource-constrained
settings. However, for detection and characterization of smaller lesions, MRI may
be superior to CECT and FDG PET scans.[20] Evidence is accumulating to show that MRI provides superior anatomic delineation,
leading to better localization of hepatic and vascular invasion.[21]
[22] MRI scans can be considered as an additional imaging modality (pre- or/and post-neoadjuvant
intent therapy) in patients with potentially resectable/addressable liver metastases
who are planned for local liver-directed therapies (LDT). Although this is not mandatory,
it may help provide better anatomical information regarding the liver lesions that
may be therapeutically relevant, may resolve diagnostic dilemmas with respect to some
indeterminate liver lesions, and also may help definitively rule out disseminated,
small metastatic lesions in the liver that would preclude local treatments (the authors recognize the results of the CAMINO trial which clearly suggest an almost
mandatory role of an MRI of the liver in patients with CRLM, though the current recommendations
were formed prior to the publication of the complete results of the trial).[23]
MRI is an important pelvic imaging modality in carcinoma rectum due to the excellent
soft tissue contrast that it provides between the tumor and other soft tissues on
T2-weighted images, which is superior to that provided by a CECT.[24]
[25] The sensitivity and specificity of MRI in rectal cancer are heavily operator dependent.[26] Parameters like level of the tumor from the anal verge, T staging, and N staging
need to be assessed for planning the type of surgery, and requirement of preoperative
chemoradiation. Assessment of risk of peritoneal perforation, feasibility of a total
mesorectal excision (TME) in middle rectal, and relationship of the tumor to puborectalis
muscle to understand the feasibility of a sphincter-preserving surgery are usually
done with an MRI of the pelvis. The high signal intensity of mesorectal fat and the
low signal intensity tumor provide a natural, intrinsic contrast in assessing the
tumor extent on T2-weighted images on MRI.[27]
[28]
[29] The excellent soft tissue resolution that MRI provides with intrinsic contrast helps
in accurate T staging that ranges anywhere between 60 and 95%, depending on the section
thickness, use of appropriate coils, and the expertise of the reporting radiologist.[28]
[29]
[30] Accurate preoperative assessment of early-stage tumors with clear distal and radial
margins helps obviate the need for irradiating the sphincter that leads to better
postoperative sphincter function and lower rates of anastomotic site breakdown.[31] In view of superior anatomical delineation, MRI of the pelvis in addition to a well-conducted
CECT of the abdomen and pelvis is recommended in those patients with either metastatic
or oligometastatic rectal cancer who are being considered for a local radiation or
a curative surgical resection of the rectal primary along with metastasectomy (in
case of oligometastatic rectal cancer).
Identification of RAS mutations forms one of the cornerstones of biomarker analysis
in the management of mCRC. Retrospective analysis and prospective trials have clearly
suggested that cetuximab (anti-EGFR-directed monoclonal antibody) had activity in
RAS wild type (RAS-WT) tumors only.[32] Similar observations were made for the use of another anti-EGFR monoclonal immunoglobulin
G2 antibody, panitumumab.[33] KRAS exon 2 activating mutations occur in approximately 37 to 45% of CRCs.[34]
[35]
[36] In addition to the ∼40 to 45% of patients with exon 2 KRAS mutations, an additional
9 to 10% of CRC patients carry alternative RAS mutations. These include 4 to 5% NRAS
mutations (codons 12 or 13 on exon 2 and codons 59 or 61 on exon 3) and another 5%
with a non-exon 2 KRAS mutation (codons 59 or 61 on exon 3 and codons 117 or 146 on
exon 4).[37] The predictive value of these “non-KRAS exon 2” RAS mutations has been evaluated
across several of the phase III cetuximab or panitumumab-based clinical trials. No
benefit was noted with the addition of either cetuximab or panitumumab to either irinotecan
or oxaliplatin-based chemotherapy across first, second, and refractory lines of therapy.[38]
[39] These results have triggered a change in practice that mandates treatment with anti-EGFR
agents to patients with extended left-sided RAS wild-type tumors only (absence of
exon 2, 3, or 4 KRAS or NRAS mutations).
BRAF mutations are an independent poor prognostic marker for metastatic colorectal
carcinoma.[40]
[41] These mutations occur in 5 to 12% of mCRC (with an increased prevalence in right-sided
mCRC), and the most common mutation is the single substitution missense mutation V600E.[42] While the negative prognostic impact of BRAF mutation has been confirmed across
various studies, debate continues regarding its value as a predictive marker of response
to anti-EGFR therapy. It has been postulated that treatment with anti-EGFR therapy
may be of limited value in these patients in view of alterations in MAPK pathway.[43] In a retrospective analysis of 113 patients treated with cetuximab or panitumumab,
11/79 (∼14%) patients of KRAS WT were found to have BRAF V600E mutations, and none
of them responded.[44] In the second-line treatment of mCRC, the PICCOLO trial reported on a subgroup of
131 patients with BRAF mutant mCRC randomized to second-line panitumumab plus irinotecan
or irinotecan alone. Patients with BRAF mutation had a trend toward a worse overall
survival with the addition of panitumumab (HR = 1.4; [0.82–2.39]).[45] On the other hand, a combined analysis of the CRYSTAL and OPUS studies showed a
nonsignificant trend in RR, progression-free survival (PFS), and overall survival
(OS) with the addition of cetuximab to FOLFIRI in BRAF mutant patients.[46] While these data are limited by the small portion of patients with BRAF mutation
in each of the above studies, they suggest a lack of a clinically meaningful improvement
in OS with the integration of anti-EGFR therapy with chemotherapy or as monotherapy
in mCRC with BRAF mutations. This makes BRAF testing essential before starting any
targeted therapy for patients with mCRC.
The landmark study by Le and colleagues, evaluating pembrolizumab (anti-PD1 antibody)
for patients with deficient mismatch repair protein (dMMR) status, showed excellent,
durable overall response rates (ORRs) in the cohort of patients with mCRC.[47] The MMR system helps heal the DNA breaks during replication; a deficiency in this
machinery causes microsatellite instability.[48] Testing for mismatch repair-deficient CRC can be conducted using IHC, PCR, or NGS.
While PCR is typically regarded as the gold standard, the concordance between PCR
and IHC exceeds 97%.[49] In practical application, IHC is commonly utilized as the standard approach. dMMR
tumors (or MSI-H) are known to have increased neoantigen load and upregulation of
tumor-infiltrating lymphocytes, thus increasing their responsiveness to immune checkpoint
blockade.[50] dMMR (or MSI-H) occurs in approximately 5% of all mCRCs. Patients with mCRC and
dMMR status have been shown to have less than expected survival with traditional chemotherapy
in comparison to patients with proficient MMR tumors.[51] Based on the landmark study, pembrolizumab initially (November 2015) received U.S.
FDA breakthrough therapy designation for the treatment of pretreated MSI-H mCRC, and
has recently shown efficacy as first-line therapy as well.[52]
One of the mechanisms by which anti-EGFR therapy fails in RAS wild-type mCRC patients
is the development of HER2 alterations (amplification or mutation) and facilitating
bypass signaling.[53] HER2 amplification is associated with poor response to anti-EGFR therapy and shorter
PFS, suggesting inherent resistance.[54] HER2 alterations occur in 4 to 5% of all mCRC patients, predominantly left-sided
tumors and are mutually exclusive with RAS mutations. The first real proof-of-concept
study was HERACLES phase II trial, which showed excellent clinical activity for dual
HER2 blockade with trastuzumab and lapatinib.[55] It is important to note that the study included strong HER2-positive tumors by taking
into consideration a higher cutoff of >50% as compared to a 10% cutoff usually taken
in breast cancers. Multiple other anti-HER2 targeting strategies have been studied
with more or less similar positive outcomes.[56]
[57] Although currently testing for HER2 in newly diagnosed mCRC is not standard of care,
with more availability of broad NGS-based testing, identifying this small definite
subgroup may be beneficial if treated with the novel targeted therapies.
Major benefits of performing NGS or CGP (comprehensive genomic profiling) over single-gene
techniques are beyond classical hotspots; we can find additional rare RAS alterations
which can directly impact our selection of anti-EGFR therapy. It also allows us to
identify non-V600E BRAF mutations, some of which might also impact our treatment decisions.[58] A very small but intriguing subset of mCRC patients have fusion genes resulting
from complex rearrangements of protein kinase.[59]
Liquid biopsy, specifically ctDNA analysis, overcomes the limitations of inadequate
tissue availability in pretreated settings, by capturing the overall genomic mutational
landscape of the disease from blood samples. ctDNA demonstrates high sensitivity (≥93%)
for detecting mutations like RAS in treatment-naive patients, even at low variant
allele frequencies (≥0.18%).[60] This is an emerging diagnostic modality which will be increasingly used in the near
future.
Cytotoxic chemotherapy remains the cornerstone of first-line systemic therapy for
the majority of patients with mCRC.
Deciding upon the chemotherapy backbone and the regimen's intensity (mono, doublet,
or triplet) involves a comprehensive evaluation of patient-related factors, including
age, performance status, comorbidities, frailty, organ function, and individual preferences
(intravenous vs. oral administration). Subsequent considerations involve disease-related
factors such as tumor sidedness, disease burden, and the rate of tumor growth, which
may impact the general condition of the patient. Additionally, treatment-related factors
such as the risk of cardiotoxicity, incidence of peripheral neuropathy, tolerance
to previous regimens, and prior exposure to adjuvant chemotherapy must be carefully
taken into account.
For nearly four decades until the late 1990s, the fluoropyrimidine drug 5-FU stood
as the sole therapeutic agent for treating mCRC. The emergence of oxaliplatin and
irinotecan, used in combination with each other and 5-FU analogs, has further increased
options and survival in this group of cancers.
Monotherapy
FOCUS4, CAIRO5, and FFCD 2000-056 were the three Phase III randomized studies that
aimed to address the question of whether sequential treatment strategies were better
than combination chemotherapies in terms of minimizing adverse events without compromising
survival.[61]
[62]
[63] Overall, these studies indicated that the sequential approach, starting with first-line
5-FU alone, is noninferior in outcomes compared to combination chemotherapy. However,
they need to be interpreted with caution as the response rates and PFS were superior
with the combination approach in all three studies, and also there was a relatively
lower exposure of patients to all three cytotoxic drugs in these studies (ranging
from 16 to 55%). Therefore, adopting a sequential strategy might result in lower disease
control, posing a risk of fewer patients being exposed to all three drugs and, consequently,
potentially lower survival rates.
In our opinion, monotherapy should be reserved for patients who are not suitable for
doublet chemotherapy due to factors such as elderly age, comorbidities, frailty, and/or
contraindications to cytotoxic agents. Addition of targeted therapy agents like bevacizumab
and anti-EGFR therapies to the single-agent chemotherapy backbone further improves
the efficacy modestly and is dealt with later.
Doublet Chemotherapy
The primary backbone for systemic chemotherapy in patients with mCRC continues to
be doublet chemotherapy. There are two main categories of regimens: those based on
oxaliplatin and those based on irinotecan, with further distinctions based on the
administration of 5-FU, either orally (via 5-FU analogs such as capecitabine, S1,
or UFT) or intravenously. Two randomized phase 3 studies have been published comparing
the FOLFOX and FOLFIRI regimens, revealing similar efficacy in terms of response rates
(around 55%), PFS (approximately 8 months), and overall survival (approximately 21
months).[64]
[65] As expected, both regimens have a differential safety and adverse event profile.
For patients who opt to avoid central venous line access or desire a longer interval
between chemotherapy, oral 5-FU analog-based chemotherapy regimens featuring capecitabine,
S1, or UFT can be considered. Numerous randomized phase 3 studies, along with a meta-analysis,
have demonstrated the comparable effectiveness and safety of capecitabine and oxaliplatin
(CAPOX/XELOX) when compared to infusional 5-FU/folinic acid plus oxaliplatin (FOLFOX).
Notably, while the FOLFOX regimen exhibited a higher incidence of neutropenia, CAPOX/XELOX
was associated with higher incidences of thrombocytopenia, diarrhea, and hand–foot
syndrome.[66]
[67]
[68]
[69]
Triplet Chemotherapy
Several studies have demonstrated the synergistic clinical activity achieved by combining
all three cytotoxic agents—fluoropyrimidine, oxaliplatin, and irinotecan.[70]
[71] The exposure of mCRC patients to this trio of agents during the course of their
treatment has been shown to provide the best chances of survival.[72] These findings prompted the development of the FOLFOXIRI regimen (folinic acid,
5-FU, oxaliplatin, and irinotecan) for the initial treatment of mCRC patients.
Two randomized studies comparing FOLFOXIRI with FOLFIRI in mCRCs showed contrasting
results: the GONO19 study showed statistical improvement in PFS and OS, while the
HORG 20 study did not.[73]
[74] Despite these discordant results, a systematic review and meta-analysis of randomized
controlled trials, comparing triplet chemotherapy to doublet chemotherapy (specifically
FOLFOX or FOLFIRI) as the initial treatment for mCRC, have revealed a relevant increase
in efficacy with an expected increase in toxicities as well.[75] There was a 25% relative risk reduction of death and a 27% relative risk reduction
in disease progression, compared with the doublet regimens. However, this improvement
came at the cost of higher rates of grade 3 adverse events.[21]
In the era of targeted therapy, the FOLFOXIRI triplet regimen has been successfully
combined with bevacizumab, further strengthening its clinical utility.[76]
Due to its synergistic activity and impressive response rates, the FOLFOXIRI triplet
regimen is the preferred choice for patients where conversion therapy is being considered
with the intention of resection, or when the disease burden is substantial, accompanied
by a rapid growth rate that threatens to deteriorate the patient's general condition.
However, it is not recommended for administration in patients older than 75 years
or those patients younger than 75 years with an ECOG PS > 0. Preferably, it should
be avoided in cases where patients have prior exposure to oxaliplatin in the adjuvant
setting.
Older patients with mCRC (> 60 years of age as per Indian criteria) should be considered
for a comprehensive geriatric assessment (CGA) if feasible before being planned for
systemic therapy. Multiple components of the CGA and the CGA itself have proven to
have correlations with adverse events as well as survival outcomes in mCRC.[77]
[78]
[79]
[80] The CGA can be used as an adjunct in planning treatment for older patients with
CRC. Additionally, randomized controlled trials in patients ≥70 years (with or without
frailty) have shown similar survival outcomes with 5-FU (on analogs) based therapy
in comparison to combination therapy, and this needs to be considered while planning
for therapy in these patients.[81]
[82]
-
3.2: First-line systemic therapy—Targeted therapy and maintenance.
The rationale for the use of targeted therapy in mCRCs, especially in RAS wild-type
tumors, has heavily depended on dividing tumors into right-sided and left-sided CRCs.
Right- and left-sided colon cancers have embryologically different origin. While the
former arises from the midgut, the latter arises from the hindgut.[83] Left-sided tumors generally have better outcomes as compared to the right-sided
ones.[84] Recent meta-analysis of 13 randomized clinical trials showed significantly worse
PFS and OS for the right-sided metastatic colon cancers irrespective of the biological
therapy.[85] These differences in outcomes can be attributed predominantly to different molecular
signatures occurring on each side. Data from The Cancer Genome Atlas (TCGA) database
and PETACC-3 adjuvant trial showed predominant epidermal growth factor receptor (EGFR)
pathway upregulation in the left-sided cancers. Even when RAS and RAF mutations were
excluded, gene signature patterns were considerably different between right and left
colon cancers.[86] A retrospective study of ∼ 2,500 CRC specimens showed statistically higher prevalence
of BRAF, PIK3CA, CTNNB1, ATM, and PTEN in right-sided colon cancer, while more patients
with left-sided primary disease had TP53 and APC gene alterations.[87]
-
3.2.1: The following targeted therapeutic options in combination with first-line chemotherapy
are considered in patients (without resource constraints) with left-sided tumors (ALL
RAS WT, BRAF WT):
-
FOLFOX plus anti-EGFR-directed antibody (1A).
-
FOLFIRI plus anti-EGFR-directed antibody (1A).
-
FOLFIRI/FOLFOX plus bevacizumab (1B).
-
FOLFOXIRI/modified FOLFIRINOX plus bevacizumab (2A).
-
CAPOX plus bevacizumab (1B).
-
CAPOX plus weekly cetuximab (3B).
-
3.2.2: The following targeted therapeutic options in combination with first-line chemotherapy
are considered in patients (with resource constraints and not feasible for anti-EGFR-directed
antibodies) with left-sided tumors (ALL RAS WT, BRAF WT):
-
FOLFIRI/FOLFOX plus bevacizumab (1A).
-
FOLFOXIRI/modified FOLFIRINOX plus bevacizumab (1A).
-
CAPOX plus bevacizumab (1A).
-
FOLFOX/FOLFIRI/CAPOX/FOLFOXIRI or modified FOLFIRINOX alone (1A).
-
3.2.3: The following targeted therapeutic options in combination with first-line chemotherapy
are considered in patients (without resource constraints) with left-sided tumors (ALL
RAS WT, BRAF WT):
-
3.2.4: The following targeted therapeutic options in combination with first-line chemotherapy
are considered in patients (without resource constraints) with right-sided tumors
and unresectable metastatic sites (not being considered for downstaging to potential
resection) of disease (ALL RAS WT):
-
FOLFIRI/FOLFOX plus bevacizumab (1A).
-
FOLFOXIRI/modified FOLFIRINOX plus bevacizumab (1A).
-
CAPOX plus bevacizumab (1A).
-
FOLFOX plus anti-EGFR-directed antibody (1C).
-
FOLFIRI plus anti-EGFR-directed antibody (1C).
Post hoc analysis of several studies showed primary tumor location to have an important
predictive effect on clinical outcomes when treated with EGFR or VEGF inhibitors.
In a subgroup analysis in the CRYSTAL and FIRE-3 study, the addition of cetuximab
to FOLFIRI had a modest additional benefit in PFS and OS (statistically not significant)
if the primary tumor was located on the right side.[88] Similar observations were made in PRIME study evaluating panitumumab with or without
FOLFOX chemotherapy. PFS and OS advantage of anti-EGFR therapy was restricted to patients
with left-sided mCRC.[89] CALGB/SWOG 80405 studied cetuximab versus bevacizumab along with doublet chemotherapy
and found poorer outcomes with cetuximab plus chemotherapy in right-sided tumors.
Although all analysis is post hoc in nature, these are clearly consistent among all
major first-line clinical trials. NCCN/ASCO recommends against the use of EGFR inhibitors
for the first-line treatment in right-sided mCRC irrespective of RAS status.
On the other hand, primary tumor location does not seem to be predictive of benefit
from bevacizumab treatment in patients with mCRC. Data from Australian prospective
multicenter mCRC registry studying 926 patients showed improved PFS with addition
of bevacizumab to chemotherapy irrespective of primary tumor location (right side:
8.5 vs. 4.9 months {HR = 0.46, p < 0.001} in favor of chemotherapy + bevacizumab (left side: 10.5 vs. 7.5 months {HR = 0.71,
p = 0.006} in favor of chemotherapy + bevacizumab).[90]
In a recent retrospective exploratory subgroup analysis of two randomized phase III
studies with bevacizumab in the first-line treatment of mCRC, Loupakis et al showed
that there was no effect of primary tumor location on efficacy of bevacizumab.[91] Therefore, bevacizumab remains targeted therapy of choice in patients with right-sided
mCRC irrespective of RAS status.
Individual patient data meta-analysis of six major randomized trials (CRYSTAL, FIRE-3,
CALGB 80405, PRIME, PEAK, and 20050181) has clearly reported a significant improvement
in RR (OR: 2.12, 95% CI: 1.77–2.55), PFS (HR: 0.78, 95% CI: 0.70–0.87, p < 0.001), and OS (HR: 0.75, 95% CI: 0.67–0.84, p < 0.001) in the anti-EGFR-containing arms.[92] Addition of bevacizumab to chemotherapy in the treatment of left-sided mCRC also
improves all efficacy endpoints; however, the magnitude of benefit is less as compared
to anti-EGFR-based therapy. Hence, one may consider using bevacizumab-based first-line
treatment for left-sided mCRC especially if the patient has concerns about skin toxicity
or costs of the treatment, but clearly anti-EGFR therapy remains the treatment of
choice for left-sided mCRC. The paradigm study, a phase III trial comparing panitumumab
to bevacizumab-based chemotherapy in predominantly left-sided, all RAS wild-type mCRC,
demonstrated a statistically significant OS advantage with panitumumab-based therapy.
However, the degree of benefit observed was significantly less compared to prior post
hoc analyses.[93]
As previously discussed in the RAS biomarkers section, anti-EGFR therapy's role in
the first-line treatment for RAS mutant mCRC is generally limited and not recommended.
However, two exceptions are currently under investigation. Firstly, the potential
of sotorasib with panitumumab is being explored for mCRC patients with the KRAS G12C
mutation, showing promise in initial case reports and phase I studies.[94] Secondly, for those with a BRAF mutation, the combination of cetuximab with encorafenib
has gained FDA approval for second-line treatment and beyond. However, its recommendation
for first-line use awaits completion of the BREAKWATER study.[95]
Bevacizumab is a recombinant humanized monoclonal antibody specifically binding and
blocking all human vascular-endothelial growth factor (VEGF)-A isoforms. In the landmark
trial AVF2107, Hurwitz and colleagues compared irinotecan, bolus fluorouracil, and
leucovorin (IFL) with or without bevacizumab in the first-line treatment of mCRC and
found significant improvement in ORR, PFS, and OS.[96] Several subsequent phase III randomized clinical trials showed consistent improvement
in PFS; however, OS was most pronounced in the setting of fluoropyrimidines monotherapy.[97]
[98]
[99] In the individual patient data meta-analysis of seven RCTs (AVF2107, NO16966, ARTIST,
AVF0780, AVF2192, AGITG MAX, and E3200), KRAS mutational status was available in 530
patients (14.1% of the total study population), which showed benefit of addition of
bevacizumab irrespective of KRAS status.[100] Therefore, bevacizumab remains targeted therapy of choice in patients with RAS mutant
mCRC.
For all practical purposes, HER2 positivity is considered for those patients who are
scored as 3+ by IHC, show HER2 amplification by fluorescent in situ hybridization
(FISH) or HER2 amplifications as detected by NGS. A majority of the treatment options
for patients with HER2-positive mCRCs have been explored in the pretreated scenario,
though upfront testing for HER2 status is encouraged by the guidelines. Trastuzumab
and deruxtecan, pertuzumab and T-DM1, and trastuzumab and tucatinib are options that
can be considered in this scenario.[55]
[57]
[101]
[102]
-
3.2.8: In patients with mCRC (RAS mutant), patients who have achieved at least stable
disease after 4 to 6 months of chemotherapy (with or without bevacizumab) can be considered
for the following:
-
3.2.9: In patients with mCRC (RAS wild type), patients who have achieved at least
stable disease after 4 to 6 months of chemotherapy (with or without cetuximab/panitumumab)
can be considered for the following:
-
Observation (1A).
-
Capecitabine or 5-FU monotherapy (1A).
-
Combination of 5 FU plus cetuximab/panitumumab(1A).
-
Cetuximab (1C).
-
Panitumumab monotherapy (1D).
Continuation of first-line chemotherapy until disease progression is an issue, especially
in the setting of oxaliplatin-based treatment in view of cumulative neurotoxicity.
Several studies have analyzed treatment breaks (also known as “treatment holidays”)
versus maintenance strategies in first-line treatment of mCRC, especially in the setting
of oxaliplatin-based doublet chemotherapy. Early indications of the feasibility of
intermittent treatment came from OPTIMOX trials, suggesting complete treatment breaks
may have inferior outcomes and the patients receiving maintenance therapy had longer
PFS.[103]
[104] In the era of biological therapy, first indications of continued maintenance therapy
came from NO16966 trial, with conflicting results from later studies.[105]
[106]
[107] In the CAIRO-3 study, patients were randomized if they had at least stable disease
after six cycles of CAPOX plus bevacizumab (CAPOX-B) to the continuation of capecitabine
with bevacizumab versus observation with the primary endpoint being PFS2 (defined
as progressive disease after re-introduction of CAPOX-B). The study achieved its primary
endpoint of improved PFS2 (8.5 months in the observation group and 11.7 months in
the maintenance group [HR: 0.67, 95% CI: 0.56–0·81, p < 0·0001]), though overall survival was only numerically improved with the maintenance
strategy.[105] The SAKK 41/06 phase III trial also suggested that observation was not noninferior
to a maintenance strategy using bevacizumab (time to tumor progression: 2.9 vs. 4.1
months, HR: 0.74 [95% CI: 0.58–0.96] in favor of maintenance therapy). The German
AIO 0207 study, another noninferior study design, evaluated three follow-up strategies
after 24 weeks of oxaliplatin-based doublet induction chemotherapy (observation, bevacizumab
alone, and bevacizumab + fluoropyrimidine) with a primary endpoint of time to failure
of strategy. Though only a small number of patients (36%) went on for re-induction
chemotherapy as compared to CAIRO 3 study (∼60%), PFS was better when both the maintenance
arms were compared to observation alone.[106]
A meta-analysis of 14 prospective studies done between 2009 and 2017 showed a small
but definite and statistically significant PFS benefit for use of maintenance therapy
versus observation, although no overall survival benefit was seen.[108]
The role of anti-EGFR therapies in the maintenance setting for mCRC has recently evolved
after retrospective studies and the results of the recently published PANAMA study.[109]
[110] The PANAMA phase II study evaluated the question of panitumumab with 5-FU versus
5-FU alone in a maintenance setting and showed an improvement in PFS, though without
significant differences in OS.[69] In conclusion, data supporting maintenance strategy in mCRC is strong for fluoropyrimidine
with or without bevacizumab or panitumumab. All patients receiving oxaliplatin-based
induction chemotherapy should be considered for maintenance therapy, although individualization
and discussion with the patient is essential, considering that “treatment holidays”
for a limited time can be acceptable for any patient with indolent and asymptomatic.
The KEYNOTE-177 trial has established pembrolizumab as the standard of care in patients
with MSI-H or dMMR mCRCs. The study met one of its primary endpoints (improvement
in PFS), though a coprimary endpoint of improved OS was not met. The lack of improvement
in OS is likely due to the high proportion of patients crossing over to the pembrolizumab
arm (60%), among other possible reasons.[111] The first results of the CheckMate 8HW study have suggested that the combination
of nivolumab plus ipilimumab also improves PFS compared to chemotherapy. Within the
confines of the early results of the CheckMate 8HW trial as well as the flaws of a
cross-trial comparison, the combination of nivolumab plus ipilimumab can be considered
as an option for the treatment of MSI-H mCRCs.[112]
Despite the available patient assistance programs for pembrolizumab and nivolumab,
these options are out of reach of the majority of Indian patients. In such a scenario,
most patients with MSI-H mCRC are started on chemotherapy with or without targeted
therapy. Based on the mechanism of action of immune checkpoint inhibitors (ICIs) and
the dosing considerations noted in early-phase studies, it is postulated that low-dose
ICIs may also have efficacy. The use of low-dose nivolumab has shown efficacy in a
randomized phase III trial in advanced head and neck cancers. Keeping the above factors
in mind, the recommendations suggest that low-dose ICIs can be considered in the Indian
scenario in MSI-H mCRCs, though there are limited data and publications with regard
to this usage.[113]
General Principles
There is significant overlap in the treatment options between patients who have synchronous
mCRC where there is a possibility of using conversion therapy and proceeding to possible
local therapy (LT), and patients with OMD.[7] Hence, for the purposes of these recommendations, both categories have been considered
together. It is also important to note that the current recommendations have limited
inputs from surgeons, interventional radiologists, radiation oncologists, and other
contributors from a potential multidisciplinary board (MTB) setup and primarily concentrates
on rationalizing systemic therapeutic options in the management of mCRC. Further updates
of these recommendations in the future will include inputs from other members on an
MTB approach. The recommendations also recognize that using a combined approach for
patients being planned for a “conversion approach” and OMD suffers from the following
drawbacks:
-
OMD is emerging as a distinct biological entity with a specific genomic profile and
is likely no longer an anatomical entity that can be considered halfway between localized
disease and diffuse metastatic disease.[114]
[115]
[116]
-
OMD is a more well-defined entity predominantly comprising resected primary lesion
(usually), and less than five sites of metastases at not more than two different organ
sites, all of which are amenable to LT. However, there are other existing definitions
and principles according to which OMD is identified and managed.
-
LT has a greater role in OMD than patients in the “conversion” pathway, with LT approaches
showing a survival benefit (predominantly DFS or PFS benefits) for using LT with or
without systemic therapy compared to systemic therapy alone.
Specific Management
A majority of evidence for the management mCRC with resectable/potentially resectable
metastases comes from the management of CRLMs. While older studies attempted to prognosticate
patients based on simple biological, anatomical, and serological criteria in patients
with liver metastases, most of these studies were conducted in an era where biomarkers
like RAS/RAF/MMR were not identified, and there were limited systemic and targeted
therapeutic options.[117] Patients with limited lung metastases as well as peritoneal-only metastases have
also come under the ambit of resection pre- or post-systemic therapy. Because of the
varying criteria used by institutions, the current recommendations suggest that all
institutions form their own guidelines for assessing resectability and LT via consensus
among treating oncologists, surgeons, radiation oncologists, intervention radiologists,
and other members of an MDT. Broadly, the following principles are suggested:
-
In patients with clearly R0-resectable primary and metastatic sites, surgery should
be considered upfront followed by systemic therapy using an adjuvant-like approach.
While there is limited evidence for using such an approach of “pseudo-adjuvant” chemotherapy,
6 months of systemic chemotherapy (FOLFOX or CAPOX) can be considered based on extrapolation
from the management of stage III CRC. The use of targeted therapy after resection
is not recommended.[118]
-
In patients where there are doubts on the possibility of R0 resection on an initial
staging evaluation, it is preferable to start with systemic chemotherapy with targeted
therapy as the initial approach, followed by 2- to 3-monthly assessments for the possibility
of resection/LT. The evidence for such an approach comes from the study by Nordlinger
et al (EORTC 40983), which compared perioperative FOLFOX4 chemotherapy to surgery
alone in patients with CRC and resectable CRLM and achieved its primary endpoint of
improved PFS in the perioperative group.[119]
-
The addition of targeted therapy as part of a “conversion” approach with neoadjuvant
intent should follow the principles as for unresectable disease, with the following
points to be kept in mind:
-
In patients with right-sided mCRC or left-sided RAS mutant/BRAF mutant CRC, using
FOLFIRINOX plus bevacizumab might be beneficial in terms of response and downstaging
compared to FOLFIRI/FOLFOX plus bevacizumab at the cost of additional treatment-related
side effects. In patients who may not be fit for a triple therapy combination, FOLFOX/FOLFIRI
plus bevacizumab is an acceptable and preferred option.[120]
[121]
-
In patients with left-sided mCRC with RAS WT/BRAF WT status, there appears to be no
significant differences in outcomes when comparing FOLFIRI/FOLFOX plus bevacizumab
and FOLFOX/FOLFIRI and cetuximab based on the results of the CAIRO5 study.[121] There is controversy with regard to the management of patients in this subset, as
previous meta-analyses have suggested increased downstaging and response rates when
anti-EGFR-directed therapy was used.[93]
[122] However, either of the two approaches (anti-VEGF-directed therapy or anti-EGFR-directed
therapy) can be used as part of conversion therapy in mCRC.
-
Assessments for resectability should be conducted by relevant imaging every 8 to 10
weeks to ensure that the window for resection is not lost and to avoid increased duration
of exposure to chemotherapeutic agents, especially oxaliplatin, which may cause increased
liver toxicity and perioperative complications.
-
Resection of the metastases should be carried out approximately 5 weeks from the previous
administration of chemotherapy–bevacizumab, assuming bevacizumab has not been omitted
from the last cycle of therapy.
-
It should be noted that resection post–systemic therapy poses a different challenge
compared to upfront resections in patients with mCRC. Use modalities such as RFA,
microwave ablation, or stereotactic body radiotherapy (SBRT) prior to resection, or
a two-staged resection may be applicable in select scenarios.[7] Portal vein embolization to ensure hypertrophy or remnant liver is also an acceptable
option in scenarios where hepatic resections are being planned in the future.
-
It is important to note that there is limited or no role of resection of the primary
tumor (barring emergent scenarios) in patients who have unresectable metastatic sites
of disease. This is now supported by a small phase III trial from Japan.[123]
A number of studies have shown a benefit for the usage of LT as opposed to systemic
therapy alone, though the quantum of benefits is not well quantified.[124]
[125] However, a majority of these studies have not accounted for disease biology via
the use of biomarkers or used monoclonal antibodies as part of systemic therapeutic
approaches. Nevertheless, as approximately 20 to 45% of patients with OMD can have
long-term survival, the use of LT should be considered paramount in the management
of OMD. A detailed description of the pros and cons of the various LTs is beyond the
scope of the current recommendations, but the recommendations do recognize the use
of modalities such as TARE, SBRT, and cytoreductive surgery (CRS), with or without
HIPEC, in the management of mCRC.[126]
[127]
At a juncture where the patient needs to be assessed for second-line chemotherapy
in mCRC, the most important factors to be taken into consideration are the ECOG performance
status of the patient and prior treatment. For patients who have progressed on oxaliplatin-based
regimens (i.e., CAPOX/FOLFOX), single-agent irinotecan has been shown to be superior
to single-agent infusional 5-FU or supportive care alone in terms of efficacy with
similar quality of life.[128]
[129] Evidence shows that addition of 5-FU to irinotecan (FOLFIRI) does not have an overall
survival advantage but offers benefit in terms of PFS and ORR with no significant
difference in toxicity. FOLFIRI can be thus offered as second-line chemotherapeutic
regimen to fit patients with a large disease burden who have previously received oxaliplatin-based
chemotherapy for metastatic disease as it gives better ORR and PFS.[130] A phase 3 study in Asian patients also demonstrated the noninferiority of mCAPIRI
as compared to FOLFIRI with or without bevacizumab in the second-line treatment of
mCRC.[131] Hence, while CAPIRI or mCAPIRI is not commonly recommended as first-line therapy
in mCRC, mCAPIRI as a second-line treatment option with reduced dose capecitabine
can be considered as an acceptable alternative to FOLFIRI without compromising the
efficacy. In patients who have received first-line FOLFIRI, superiority of FOLFOX
over single-agent 5-FU or oxaliplatin was seen in phase 3 RCT in mCRC patients, where
it offers a clinically relevant and statistically significant PFS advantage.[132] CAPOX and FOLFOX regimens can be used interchangeably based on the clinical scenario
and patient profile.[133] Oxaliplatin monotherapy is not commonly used as a second-line agent in mCRC.
In the second-line treatment of mCRC, whether or not bevacizumab has been used in
the first-line setting, it may be added to any second-line chemotherapy regimen based
on evidence of randomized phase 2 and phase 3 studies as well as their pooled meta-analysis
that showed a modest but clinically meaningful and statistically significant survival
improvement.[134]
[135]
[136]
[137] Evidence from a randomized controlled trial suggests a bevacizumab dose of 5 mg/kg
body weight to be equally efficacious as 10 mg/kg body weight in second-line mCRC
patients who have progressed on oxaliplatin-based therapy.[138] This is particularly relevant to resource-constrained settings where ensuring lowered
costs helps more patients benefit from these drugs. Although there is no randomized
trial comparing single-agent irinotecan with irinotecan plus bevacizumab in second-line
mCRC treatment, in light of the evidence from other trials previously quoted, it is
a reasonable treatment option.
Based on evidence from a randomized phase III study that showed a statistically significant
overall survival benefit, ramucirumab can be combined with second-line FOLFIRI in
patients who have progressed in first-line chemotherapy with or without bevacizumab.[139] The risk of arterial and venous thromboembolic events and life-threatening gastrointestinal
or nongastrointestinal bleeding is not increased by the addition of ramucirumab, unlike
the other antiangiogenic agents, making it a relatively safe drug.[140] Data are lacking regarding the benefit of the addition of ramucirumab to FOLFOX
in second-line mCRC treatment. However, we can use the combination of FOLFOX + ramucirumab
in the second-line if the patient has received irinotecan-based therapy in the first-line.
-
5.3: The following targeted agents can be considered as second-line therapy in combination
with chemotherapy in patients with RAS wild-type mCRC (irrespective of tumor sidedness
and assuming they have not received anti-EGFR monoclonal antibodies during initial
therapy):
-
Cetuximab (2B).
-
Panitumumab (2B).
-
5.4: The following targeted agents can be considered as second-line therapy in combination
with chemotherapy in patients with RAS wild-type mCRC (assuming they have received
anti-EGFR monoclonal antibodies during initial therapy):
-
Cetuximab (2C).
-
Panitumumab (2C).
Based on randomized controlled trial data that showed PFS benefit and better ORRs
with the addition of panitumumab to FOLFIRI in second-line RAS wild-type mCRC patients,
panitumumab or cetuximab can be added to FOLFIRI/irinotecan in second-line mCRC therapy
in patients who have not received EGFR-targeted therapy in the first line. These combinations
have shown limited benefits in PFS without improvements in OS.[45]
[141]
[142] Data are lacking regarding the benefit of the addition of panitumumab or cetuximab
to FOLFOX in second-line mCRC treatment. However, we can use the combination of FOLFOX + panitumumab
in the second line if the patient has received irinotecan-based therapy in the first
line. In patients who have already received EGFR-targeted therapy in the first line,
there are limited data to support the continuation of the EGFR-targeted agent along
with a change in the chemotherapy backbone.[143] Under the selection pressure of anti-EGFR therapies, CRCs that are RAS wild type
may become resistant to EGFR-targeted therapies by altering the RAS mutation status.
Additionally, due to tumor heterogeneity, both RAS wild-type and RAS mutated clones
may co-exist from the baseline. Once the RAS WT tumor cell population reduces due
to the effect of EGFR-targeted therapies, the RAS mutated cells proliferate and repopulate
the tumor, making the tumor resistant to EGFR-targeted therapies. Also, RAS mutations
can develop de novo in previously RAS WT tumors, especially mutations in the extracellular
domain of the EGFR.[55] Both these mechanisms contribute to the development of resistance to EGFR-targeted
therapies when used in the first line. Hence, the belief that patients with RAS WT
mCRC who have had progression on first-line treatment, which includes EGFR-targeted
agents, will become permanently resistant to further EGFR-targeted therapy may not
be completely true.
Sometimes, sensitivity to EGFR-targeted therapies may be restored by a treatment break.
This forms the rationale for continuing the EGFR-targeted therapy and changing the
chemotherapy backbone alone.[144]
[145]
[146]
[147]
[148] However, in view of the small sample sizes of these studies, this is usually not
a recommended approach.
-
5.5: The following treatment options can be considered in patients with mCRC (irrespective
of RAS status), refractory to oxaliplatin and irinotecan-based therapy.
-
Regorafenib full dose (1A).
-
Regorafenib via dose escalation approach (2A).
-
TAS-102 (1B).
-
TAS-102 plus bevacizumab (1A).
-
5.6: The following treatment options can be considered in patients (without resource
constraints) and with mCRC (RAS WT) refractory to oxaliplatin- and irinotecan-based
therapy.
-
5.7: The following treatment options can be considered in patients (with resource
constraints) and with mCRC (RAS WT) refractory to oxaliplatin- and irinotecan-based
therapy.
-
5.8: Rechallenge with chemotherapy can be considered post two lines of chemotherapy
(oxaliplatin and irinotecan based) in a select group of patients in mCRC.
Regorafenib is an oral, small molecule inhibitor of multiple kinases. Two large, randomized
phase III trials in mCRC patients who have progressed on two or more lines of therapy
including 5-fluoropyrimidines, oxaliplatin, irinotecan, and monoclonal antibodies
have demonstrated the benefit of regorafenib as compared to placebo in terms of PFS
and OS. The benefit was modest but statistically significant.[149]
[150] The phase II REDOS trial explored the approach of dose escalation strategy for regorafenib
by starting at a lower dose of 80 mg per day and gradually increasing the dose.[151] This approach was associated with lower incidence of toxicity, especially the HFSR,
without compromising the efficacy of regorafenib. Hence, this dose escalation approach
may be used as a reasonable alternative to standard dosing of regorafenib, among other
approaches.[152]
[153]
TAS-102 is an oral fixed drug combination of the thymidine analog trifluridine and
a thymidine phosphorylase inhibitor tipiracil. Tipiracil prevents the quick degradation
of trifluridine. TAS-102 as monotherapy or in combination with bevacizumab has shown
modestly improved survival compared to placebo alone.[154]
[155]
[156] Like regorafenib, TAS-102 (with or without bevacizumab) can be considered as an
option in chemotherapy-refractory mCRC.
In patients refractory to irinotecan- and oxaliplatin-based therapies, both panitumumab
and cetuximab can be used as monotherapy in patients with RAS WT tumors based on modest
benefits in survival in this setting in prospective trials.[157]
[158]
[159] Both the EGFR-targeted agents cetuximab and panitumumab provide superior PFS and
OS as compared to best supportive care alone in chemotherapy-refractory mCRC patients
and can be considered, especially in patients who have never received any EGFR-targeted
therapies in the earlier lines and have a good performance status. However, in resource-constrained
settings, the quantum of benefits seen with using anti-EGFR agents as monotherapy
is not preferred. In such a resource-constrained scenario, re-challenge with the initial
chemotherapy regimen that was used is an option, though backed only by weak evidence.[160]
[161] Assessment of residual toxicity from previous chemotherapy drugs must be done before
rechallenging, especially with issues like neuropathy due to oxaliplatin.
In patients with chemotherapy-refractory BRAF-mutant tumors, a combination of the
BRAF inhibitor encorafenib and cetuximab leads to superior OS and ORR as compared
to cetuximab with irinotecan or FOLFIRI.[162] Hence, this regimen is strongly recommended for this subset of mCRC patients, though
procurement of encorafenib in India is difficult.
-
6. Follow-up protocols
There are no firm guidelines for the frequency of follow-up evaluations and examinations
of patients during the course of their treatment for mCRCs. While patients on active
therapeutic interventions will be on frequent evaluation, assessment of radiological
and serological disease status using carcinoembryonic antigen (CEA) antigen levels
with or without cancer antigen (19.9 (CA 19.9) can be considered every 8 to 12 weeks,
though there are no fixed criteria for the same. While CEA is a useful tool for monitoring
on follow-up, there are a number of nonmalignant causes for rise in CEA and these
should be kept in mind before acting on minor perturbations in these values.[163]
[164] A majority of clinical trials consider 6 to 8 weekly tumor-related assessments including
CEA and radiology, though such frequent assessments may not be required in clinical
practice unless there are particular patient-specific scenarios.
-
7. Principles of management not covered in the statements
Certain important aspects of the management of mCRCs were not covered in the discussions
during the consensus statements. These will be taken up in future versions of the
statements, but have now been mentioned in brief.
-
7.1: The sensitivity and specificity of CEA and CA 19.9 levels during diagnosis and
monitoring in patients with mCRC had not been discussed during the formulation of
the consensus statements. A detailed review will be considered in future revisions
of these statements.
-
7.2: DPD genotype testing was not specifically discussed during the development of
the consensus statements. However, based on the FDA guidelines as well as guidelines
detailed by the Clinical Pharmacogenetics Implementation Consortium (CPIC), the discussants
broadly agreed that the pros and cons of testing for DPD genotypes should be discussed
with patients being planned for 5-FU-based (or analogs of 5-FU) systemic therapy while
being counseled for treatment. Capecitabine or fluorouracil should not be recommended
for use in patients known to have clinically relevant and proven homozygous or compound
heterozygous DPYD variants that result in complete DPD deficiency. With regard to
partial DPD deficiency in those genotypes known to cause 5-FU (or analogs of 5-FU)
related adverse events, dose modifications can be considered, though there is limited
consensus on the same.[165]
[166] Available Indian data across cancers have shown significant variability in the prevalence
of heterozygous and homozygous DPD genotypes and needs further systematic evaluation.[167]
[168]
-
7.3: The appropriate duration for follow-up of patients not on active cancer-directed
therapy is a matter that requires further evaluation in the Indian context. CT scans
or PET scans are commonly used every 8 to 12 weeks, though there is limited evidence
to support for or against this practice. This topic will be considered for evaluation
in subsequent revisions of the guidelines.
