See example code and modify according to your data:
%% Make test data
fig = figure;
ax = axes(fig);
nPoint = 1001;
nSeries = 5;
k = ones(nSeries,1) * linspace(0,100,nPoint);
theta = randn(nSeries,1);
% Toy x-data
x = cos(k.*2*pi*0.1+2*pi.*theta).^(randi(4,5,1)).*(randn(nSeries,1) * 10);
% Toy y-data
y = randn(nSeries,nPoint) .* (randn(nSeries,1) * 10);
% Toy z-data
z = sin(k.*2*pi*0.1 + 2*pi.*theta).*(randn(nSeries,1) * 10);
% Labels
lab = cell(nSeries,1);
for i = 1:nSeries
lab{i} = sprintf('TestData %g',i);
end
% Plot them
plot3(ax,x.',y.',z.');
legend(lab);
%% Get 3D data
% Can use 'gca' here if don't have 'ax' handle
c = findobj(ax,'Type','line');
% c = findobj(gca,'Type','line'); % should work for your data
% This returns data in cell array, where each element matches
X = get(c,'XData'); % X{1,1} corresponds to Y{1,1} and Z{1,1}
Y = get(c,'YData'); % Y{2,1} corresponds to X{2,1} and Z{2,1} etc
Z = get(c,'ZData');
