Keywords
metastatic colorectal cancer - long-term survivors - chemotherapy - India
Atul Sharma
Introduction
As per Globocan 2018 data colorectal cancer (CRC) is the third most common cancer
(10.2%) and the second most common cause of cancer-related death among all cancer
cases.[1] Metastatic colorectal cancer (mCRC) has dismal outcomes, with the 5-year survival
rate of less than 10%. The median survival time of patients with untreated mCRC is
approximately 5 months.[2] The goal of treatment in majority of cases remain palliative unless there is limited
metastatic disease which is resectable, as surgery is the only curative treatment
option.[3] However, the landscape of treatment of mCRC has significantly evolved over the past
two decades with the availability of newer chemotherapeutic, targeted agents.[4] A better understanding of the administration of combination chemotherapy (oxaliplatin
and irinotecan) with use of biological agents (bevacizumab, cetuximab, and panitumumab),
molecular biology of the disease, approval of newer agents (regorafenib, Her2neu inhibitor)
immune checkpoint inhibitors, and improvement in supportive care delivery have largely
revolutionized the outcomes of metastatic patients.[5] A small subset of these patients may have durable responses and survival even without
surgery. The data regarding this subset are lacking from the Indian as well as South
East Asian subcontinent regarding long-term survivors. Here we report a case series
from India, of long-term survivors of mCRC mainly treated with chemotherapy.
Methods
This is a single-center retrospective analysis of patients of mCRC registered at Institute
Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi, from the
year 2005 to 2015. Main inclusion criteria were mCRC patients who were treated at
the centre mainly with chemotherapy, who had either initial presentation as metastatic
disease or those who progressed after initial surgery with or without adjuvant therapy,
and those who survived for 5 or more years. Data were retrieved from the hospital-based
record system. Demographic details including age, sex, Eastern cooperative oncology
group (ECOG)performance status, primary and metastatic sites stage, and various treatment
lines including biologics received by the patient were documented in a structured
proforma. Treatment response and current status of the patient were obtained telephonically
and from case records. Common terminology criteria for adverse events (CTCAE) 4.0
was used to document toxicities. Median and range were used to describe quantitative
variables. Kaplan–Meier curve was used for survival analyses. Cox proportional hazards
model was used to assess the association of clinical variables with survival. P values of <0.05 were considered statistically significant. Statistical analysis was
done using STATA software (ver13 StataCorp, College Station, Texas, United States).
Data were censored on September 30, 2020.
Results
Patient Characteristics
In the duration extending from year 2005 to 2015, records of 370 mCRC patients were
identified and reviewed. Thirty-eight patients fulfilled the criteria of survival
beyond 5 years. However, details of treatment, evaluable response, and complete records
were available for 31 patients. Median age was 53 years (range, 22–74 years). Sixteen
patients were women (51.6%). Out of the 31 patients, 13 (41.9%) patients had comorbidities
like diabetes, hypertension, or hypothyroidism. Twenty-eight (90.3%) patients had
ECOG PS 1 and two (6.4%) patients presented with PS 2. Twenty-one (67.7%) patients
had left-sided colon cancer, whereas 5 patients (16.1%) had right-sided colon cancer.
Five (16.1%) patients had involvement of transverse colon. The commonest histopathology
grade was moderately differentiated adenocarcinoma (87%); only one patient had mucinous
histology. Liver was the most common predominant site of metastasis (n = 18, 58%), followed by peritoneum (n = 6, 19.3%), and lung (n = 4, 12.9%). Multiple unresectable liver metastases (defined as >3) were seen in
6 patients. Two patients had family history of gastrointestinal (GI) cancers without
further details. Baseline median CEA (carcino embryonic antigen) was 6 ng/mL (range
2-900). RAS status was not available in all cases. K RAS status was available for
15 patients NRAS available in 3 patients and BRAF in one. Eleven patients were RAS
wild type whereas one patient was documented as RAS mutated ([Table1]).
Table 1
Patient characteristics
|
Characteristics
|
N = 31
(N)%
|
|
Median age
|
53 (22–74)
|
|
Sex
|
|
|
Male
|
15 (48.39)
|
|
Female
|
16 (51.61)
|
|
ECOG PS
|
|
|
0
|
1 (3.23)
|
|
1
|
28 (90.32)
|
|
2
|
2 (6.45)
|
|
Sidedness
|
|
|
Right
|
5 (16.13)
|
|
Left
|
21 (67.74)
|
|
Transverse colon
|
5 (16.13)
|
|
Mutation status available
|
|
|
K RAS
|
15 (48.38)
|
|
N RAS
|
3 (9.67)
|
|
BRAF
|
1 (3.2)
|
|
No. of previous lines received
|
|
|
1
|
14 (45.16)
|
|
2
|
8 (25.81)
|
|
3
|
3 (9.68)
|
|
4
|
1 (3.23)
|
|
5
|
5 (16.13)
|
|
No. of metastatic sites
|
|
|
1
|
27 (87.1)
|
|
2
|
4 (12.9)
|
|
Site of metastasis
|
|
|
Liver
|
18 (58.06)
|
|
Lung
|
4 (12.9)
|
|
Peritoneum
|
6 (19.35)
|
|
Distant nodal
|
5 (16.13)
|
|
Biochemical parameters mean (range)
|
|
|
CEA levels (ng/dL)
|
6 (2–900)
|
|
Hemoglobin (g/dL)
|
11.7 (6–16.4)
|
|
Total leucocyte count (109/L)
|
7,600 (4,500–15,500)
|
|
Platelets (109/L)
|
2.47 (1.01–6.09)
|
|
Serum albumin (g/dL)
|
3.9 (2.7–5.3)
|
Treatment Characteristics
Twenty-four (77%) of the patients presented as upfront metastatic disease. Seven (22.5%)
progressed as metastatic disease following previous curative resection with or without
chemotherapy and radiotherapy. Median time to relapse was 484 days. Three patients
were stage II and did not receive any adjuvant therapy. Remaining three patients received
modified folinic acid, oxaliplatin, fluorouracil–based combination (FOLFOX) (n = 3) and one patient (rectal cancer) received adjuvant Chemoradiation. Twenty patients
(64.5) received oxaliplatin-based first-line therapy. The most common regimen used
as first line in palliative metastatic setting was capecitabine and oxaliplatin (CAPOX)
(n = 11, 35.48%) followed by mFOLFOX (n = 9, 29.03%) and folinic acid, irinotecan, fluorouracil–based combination (FOLFIRI)
(n = 5, 16.13%). Only three patients underwent surgery for metastatic disease (pulmonary
metastetectomy and hepatic resection.). The number of lines of chemotherapy given
in metastatic setting were one (n = 14, 45.16%), two (n = 8, 25.81%), three (n = 3, 9.68%), four (n = 1, 3.23%), and five (n = 5, 13%). Total of 14 patients received monoclonal antibodies of which 11 patients
received in first line, 10 in second line, and 5 in third line setting. The most common
biologic agent used was cetuximab (n = 12) followed by bevacizumab (n = 1) and panitumumab (only 1). Three patients received both anti–epidermal growth
factor receptor (EGFR) and anti–vascular endothelial growth factor (VEGF) targeting
biological therapies.
Adverse Events
The most common adverse event profile CTCAE) grades 3 and 4 from long-term exposure
to multiple lines of chemotherapy were hematological toxicity (n = 8) and peripheral neuropathy (n = 2). Two patients had grades 3 and 4 diarrhea and febrile neutropenia. Other less
common grades 3 and 4 adverse events included rash and hand foot syndrome (n = 1) patient. No death was reported due to toxicity.
Outcomes
At the time of analysis, 10 patients are currently surviving. Six patients are maintaining
radiological complete response. Three patients were lost to follow-up. Median overall
survival (OS) of this cohort is 81.6 months (95% confidence interval [CI], 69.73–117.9)
([Fig. 1]) and median progression-free survival (PFS) to first line therapy was 17.23 months
(95% CI 13–46.8).
Fig. 1 Overall survival of metastatic colorectal cancers (long term survivors).
Factors Predicting Survival
We performed regression analysis to determine the factors that predicted survival.
The variables considered in our study were age, sex, sidedness, number of chemotherapy
lines received, upfront metastatic presentation, and use of biologic agents. Female
gender was the only factor found significant (p = 0.031) in univariate analysis; however, the significance was not seen on multivariate
regression analysis ([Table 2]).
Table 2
Treatment characteristics
|
Characteristics
|
(N)%
|
|
Chemotherapy regimens in first line
|
|
|
FOLFOX
|
9 (29.03)
|
|
FOLFIRI
|
5 (16.13)
|
|
Cap-OX
|
11 (35.48)
|
|
Cap IRI
|
2 (6.45)
|
|
IFL
|
4 (12.9)
|
|
Biological agents received
|
|
|
Yes
|
14 (45.16)
|
|
No
|
17 (54.8)
|
|
Progression free survival (first line)
|
17.23 months(13.0–46.8)
|
|
Median overall survival
|
81.6 months(69.73–117.9)
|
Table 3
Factors predicting overall survival
|
Factors
|
Univariate (p-value)
|
Multivariate (p-value)
|
|
Age (57 or less vs. more than 57)
|
0.981
|
–
|
|
Sex
|
0.031
|
0.174
|
|
Sidedness
|
0.244
|
–
|
|
Upfront metastatic presentation
|
0.48
|
–
|
|
Number of lines used (< or equal to 2)
|
0.084
|
0.721
|
|
Use of biologics
|
0.057
|
0.148
|
Discussion
Long-term survivors of mCRC in literature has been defined conventionally as survival
more than 5 years. mCRC has a median survival of 6 to 10 months with time-tested single
agents like 5 FU.[6] Advent of doublet chemotherapeutic regimen including oxaliplatin and irinotecan
has further contributed to the armamentarium against this disease.[7]
[8] The new era of biologicals and targeted therapy has ushered in changes in the outcomes
and there is still uncertainty on how and to what extent these agents have an impact
on survival. In the metastatic setting, “cure” per se has been a rare occurrence in
colorectal cancers. However, in recent times long-term follow ups of various large
multicenter trials have demonstrated a significant rise in patient population who
have survived beyond the 5-year mark.[9]
[10] Evolution in surgical techniques like hepatic resection of metastasis has also contributed
to prolongation in survival as reported in various studies.[11] Addressing the most common site of metastasis with improved infusional chemotherapy
as well as newer surgical techniques has resulted in improvement of disease outcomes.[12]
[13] Concrete data to substantiate correlations of factors to long-term survival and
actual magnitude of patient population who are achieving cure in real life is not
much reported. Till date, only two large studies, MD Anderson Cancer Center[9] and NCCTG,[10] have addressed these patients and their characteristics. No data from India cohort
are currently available which can provide us with the prevalence of prolonged survivors
in our population.
In our study, 77% of patients presented to us as upfront metastatic disease. This
analysis demonstrates that only a small proportion of our patients (10%) underwent
metastatectomy and majority received chemotherapy only as part of therapy. Hence,
our analysis highlights good and durable response achieved with chemotherapy alone.
Presently six patients are maintaining complete response. This is somewhat comparable
to the large retrospective study of mCRC by Ferrarrato et al from MD Anderson Cancer
Centre,[9] where they reported six patients in CR on chemotherapy alone. Majority of our patients
presented with left-sided colon cancer which now we know has a better prognosis.[14] The most common used biologic in our study was cetuximab. However, as the current
evidence of use of cetuximab according to sidedness of the colon cancer was not available
at the time of institution of biologics in our patients, the response correlation
according to primary tumor location needs to be assessed in a larger cohort.[15]
Toxicity is also a concern in long-term survivors and adverse effects like neuropathy
can be disabling and have a bearing on quality of life.[16] Seven percent of our patients had residual grade III peripheral neuropathy.[17] Quality of life also takes precedence in analysis of long-term survivors and very
few studies are available in literature to highlight this aspect. Our study did not
include quality of life analysis.
It is hard to overlook the concepts of precision medicine and tailoring of therapy
in subsequent lines. The approved agents like bevacizumab, cetuximab, and panitumumab
have shown promise in prolongation of survival.[18] Our study almost 50% patients received at least one biologic agent. Genetic determinants
of prolonged survival can be further analyzed in retrospective manner from bedside
to the bench by tissue preservation and subjecting sample to specific of molecular
and genetic testing.
The limitations of our study is its retrospective nature, less number of patients
(n = 31), limited number of patients exposed to all available options, and unavailability
of MSI (microsattelite instability) and RAS (RA\at sarcoma virus gene) status in majority
of the patients.
Survival data on newer regimens are sparse owing to heterogeneous nature or retrospective
nature of studies. Addition of biologics may have a contribution to patients achieving
and maintaining complete and partial responses beyond 5 years and needs to be examined
in detail.
Conclusion
The term “cure” might not be a remote possibility in mCRC. A close scrutiny needs
to be maintained on all the large multicenter trials and their long-term follow-ups
to ascertain and validate these data. Although the patient number is small, durable
responses achieved with chemotherapy and biological agents alone in a metastatic setting
is encouraging and possible. Analysis from large institutional database should be
carried out.
