3.2 RELATIONSHIP BETWEEN EXPOSURE TO OPIOIDS AND TUMOR RESPONSE
Spearman’s Rank Correlation coefficient was performed to determine the
strength and direction of the linear relationship between the total
weight-based morphine equivalents and tumor response.
The correlation coefficient of
-0.0103 and P-value of 0.9525 indicates a statistically insignificant,
weak negative relationship between total weight-based morphine
equivalents and tumor volume ratio (Figure 1). The correlation
coefficient of 0.1096 and p-value 0.5247 indicates statistically
insignificant weak positive relationship between total weight-based
morphine equivalent and Curie score (Figure 2).
DISCUSSION
The impact of opioids on cancer outcomes has been an ongoing topic of
considerable debate and investigation.22-26Cell-mediated immunity is critical for the eradication of tumor cells
and prevention of metastases development.27,28Multiple preclinical studies have demonstrated the suppressive effects
of opioids on the immune system in general and NK cells
specifically,18,22-25,29 but this effect is variable
amongst different types of opioids.22 Clinical studies
have been equivocal and results are inconsistent across different types
of cancers.2,30,31 This observation likely suggests
that the impact (if any), is unique to certain subtypes of cancer and
cannot be applied to cancer generally. For this reason, it is important
to investigate the impact of opioids on cancer outcomes within specific
diagnoses, such as NB. It is known that NB treatment with anti-GD2 mAb
relies on NK cells as the effector cells of the tumor cell
killing.6,7 Therefore, one would expect that increased
opioid consumption would correlate with decreased tumor killing.
Our findings did not conclusively support the general hypothesis that
the magnitude of opioid exposure negatively impacts anti-GD2
mAb-directed tumor reduction. However, this does not indicate that there
is no correlation, as we did find a weakly negative correlation (as
hypothesized) between primary tumor volume reduction and total
weight-based morphine equivalents. Interestingly, we found the opposite
correlation (weakly positive) between tumor reduction outside the
primary tumor (as measured by Curie score) and total weight-based
morphine equivalents. This was the opposite of what was hypothesized.
However, neither measure achieved statistical significance. It is likely
that the impact on tumor reduction may be too minimal to be detected by
our study methodology. While the study design is novel (the use of tumor
reduction as a marker of opioid impact on chemotherapy effectiveness),
it may not be sensitive enough. If opioids suppress NK cell activity in
vivo, as seen in vitro, the clinical significance is likely too minimal
to be detected by tumor volume reduction at the primary tumor site or
other locations. The current anti-tumor therapies may be effective
enough that small variations in the tumor response in relationship to
opioid exposure may not be detected by this study design and with this
small sample size. Still, this study design is unique in that it
investigated the impact of more than just a single perioperative opioid
exposure, as is the design seen in most investigations of the impact of
opioid/anesthesia on cancer outcomes.30-33 While the
opioid exposure in this study was more than perioperative, it can still
be classified as an acute opioid exposure (8 total days). It is
possible that since there is a different mechanism by which acute versus
chronic opioid administration modulates the immune
system,15,16 the duration of opioid therapy in our
patient population did not have as significant an impact on their tumor
progression as a patient population with a more chronic exposure to
opioid therapy would have.
Not only does the impact of opioids on cancer likely vary with duration
of opioid exposure, but also opioid dose. Studies have shown that higher
doses of morphine affect NK cells differently than do lower doses.
Higher doses lose specificity for MOR and can bind to delta opioid
receptors, which may increase NK cell
activity.15,16
A potential limitation of this study may be the lack of accounting for
differences in immunosuppression by opioid type , as all opioid
doses were converted to morphine equivalent doses. Justifiably, this
choice was based on the small number of patients and occurrences of pain
treatment with opioids other than morphine. It is known that different
drugs within the opioid class inhibit NK cells to different
degrees16,22 and our study was not powered enough to
detect differences based on the type of opioid used. Finally, another
potential limitation of this study is related to the role of
interleukin-2 (IL-2). All patients on the institutional protocol
received IL-2 on days following the anti-GD2 mAb infusion (following
course 1 and course 2), which has a pro-NK cell
effect.34 It is conceivable that the pro-NK cell
effect of the IL-2 counteracted or masked any potential anti-NK cell
effect of the opioid.
Although our results do not suggest a clinically significant correlation
between opioids converted to morphine equivalent doses and adverse
effects on tumor reduction, it should not be presumed that opioids can
be used without limitation. The warnings from preclinical studies that
demonstrate the immunosuppressive effects of opioids can be heeded
without compromising patient care. It may be prudent to favor the use of
opioids that have been shown to cause lesser degrees of
immunosuppression (e.g., hydromorphone, oxycodone,
tramadol).16,22 Additionally, it is advisable to
maximize the use of non-opioid analgesics to minimize opioid
consumption, particularly if a non-opioid analgesic such as lidocaine
has been shown to have anti-tumor effects.35-38 This
approach to the management of pain associated with anti-GD2 mAb has been
demonstrated previously.12 Gorges et al. reported that
dexmedetomidine and hydromorphone (both drugs that are not known to
affect NK cells) can be safely and effectively used to treat pain in
this patient population.12 Regardless, the effective
treatment of pain is essential as untreated pain shows similar
immunosuppressive effects.39,40
In conclusion, our study did not
find a statistically significant correlation between the consumption of
opioids and the NK cell mediated killing of NB cells as measured by
effects on tumor volume/tumor load. Although our findings do not endorse
indiscriminate or excessive opioid utilization, it is reassuring that
the doses used in this study to ameliorate the severe pain that
accompanies antibody treatment do not have adverse effects on tumor
response to antibody. Opioid sparing measures can and should be
employed, when possible, given the untoward secondary effects of
prolonged opioid use, regardless of any potentially negative influence
on cancer biology.
CONFLICTS OF INTERESTS
The authors report no conflicting interests.
ACKNOWLEDGEMENTS
The authors thank Vani Shanker, PhD, ELS for scientific editing of the
manuscript.
FUNDING
National Cancer Institute Cancer Center Support Core Grant; Grant
Number: 2P30CA021765; ALSAC
The content is solely the responsibility of the authors and does not
necessarily represent the official views of the National Institutes of
Health.
DATA AVAILABILITY STATEMENT: The data that support the
findings of this study are available from the corresponding author
upon reasonable request.