Page 35 - Read Online

Udukala et al. J Cancer Metastasis Treat 2020;6:25 I http://dx.doi.org/10.20517/2394-4722.2020.45 Page 9 of 13

However, it should be noted that for all three MMPs, their expression levels in the sera of NSCLC patients
at stage 1 and the healthy control group were statistically different.

In opposite to all other protease activities studied here in the serum of NSCLC patients, MMP13 was found
to be unsuited for both early cancer diagnosis and NSCLC staging. As shown in Figure 7, MMP13 activity
was basically the same for healthy subjects and NSCLC patients of all three investigated stages. This finding
[39]
was unexpected, because MMP13 is known to be involved in endothelial-mesenchymal transition .
DISCUSSION
The sensitivity of the Fe/Fe O nanoparticle-based fluorescence nanobiosensors permits the accurate
3
4
[5]
measurement of the activities of nine signature proteases in serum samples (30 μL) obtained from
NSCLC patients. This technology permits the rapid and inexpensive detection of NSCLC at stage 1 by
means of a simple liquid biopsy. We estimate costs of approx. $20 for measuring the activity of the 8
required proteases in serum. Six proteases permit the detection of NSCLC at stage 1. MMP1 is the best
-9
candidate for the detection of NSCLC, due to the large increase in activity of 3.16 × 10 mol/L of stage 1
-15
patients compared to 1.35 × 10 mol/L the control group. Principally, a protease-based liquid biopsy for
NCSLS has the potential of significantly reducing lung cancer mortality, because lung cancer treatment
would be more successful when the cancer would be detected at stage 1 instead of stages 2 and 3, which is
usually the case at present day. It should be noted that this technology works for the detection of virtually
all solid tumors, of which many feature distinct protease signatures. When comparing the protease
signatures of breast cancer and NSCLC, there are similarities and differences: (1) for all investigated
[5]
proteases, increased activities are detected for consecutively higher cancer stages, compared to the control
group of healthy subjects; (2) Cathepsin B is more significant for breast cancer than for NSCLC; (3) on
the other hand, MMP7 is far more sensitive in detecting NSCLC than breast cancer; and (4) MMP13 is an
unsuitable biomarker for cancer detection in either case. Based on this comparison, it is our conclusion
that the similarities between the protease expression pattern of NSCLS and breast cancer far outweigh the
differences. Therefore, the panel of proteases should be expanded to detect characteristic differences, which
could be used for the identification of the type (and the stage) of solid tumor by means of a liquid biopsy.
The latter should be offered in regular intervals to all members of cancer risk groups.

The “Significance Table 2” summarizes our findings. It shows 95% confidence intervals and P-values
[33]
calculated for the comparison of the members of individual NSCLC stages with the group of healthy
volunteers. P-values < 0.05 are considered significant and shown in green. Cathepsin L, and MMP1, 2, 3,
7, and 9 permit the detection of NSCLC at stage 1. The statistically non-significant P-values are marked in
red.

In Table 3, the average activities for all nine investigated proteases are summarized. The calibration curves
[5]
in the presence of human serum that are discussed in reference were run again in parallel to the protease
measurements described here. They were used to calculate the protease activities for NSCLC patients and
the control group of healthy volunteers. To date, most of the protease measurements in cancer research
and clinical diagnosis are performed by means of immunoassays . Whereas the latter measure the total
[10]
concentration of protease, our optical nanobiosensors determine only the fraction of active proteases.
As discussed earlier, a complex protease network exists in human biology that has the ability to form
activation cascades [20,31] . Furthermore, protease zymogens can act as signaling peptides, depending on
their glycosylation pattern . Comparing the concentrations of active and inactive proteases in cancer and
[40]
numerous other diseases may offer an unprecedented insight into the human proteasome and also provide
diagnostic opportunities.